Urban trunk road signal control coordination optimization method considering multiple types of variable lanes

Abstract

The invention relates to an urban trunk road signal control coordination optimization method considering multiple types of variable lanes, and the method comprises the steps: constructing an accumulated arrival curve and an accumulated departure curve of the peripheral section of an urban trunk road based on an accumulated curve theory and collected data, and considering the difference distribution of driving behaviors, constructing a virtual cumulative departure curve of the peripheral section of the urban trunk road; constructing a multi-objective optimization function based on the relationship among the three curves, respectively calculating the queuing distribution of the peripheral section and the internal section of the urban trunk road, and quantitatively representing the multi-objective optimization function; a constraint condition of a multi-objective optimization function is constructed by considering a vehicle conflict phenomenon, a lane number, traffic volume fluctuation, an urban trunk road public period length and traffic light duration; and solving the multi-objective optimization function based on a particle swarm optimization algorithm, and obtaining a signal control scheme and a lane division scheme. Compared with the prior art, the method can achieve the coordination optimization of urban trunk roads and signal control under the condition of multiple types of variable lanes.

Description

TECHNICAL FIELD

[0001] The present application relates to the field of traffic information, in particular to a method for coordinating and optimizing signal control of urban trunk roads considering multiple types of variable lanes. BACKGROUND

[0002] The root cause of traffic problems lies in the imbalance between traffic supply and demand. At present, scholars mainly optimize signal control parameters, such as public cycle length, green ratio, coordinated phase difference, or lane functions, such as tidal lanes and variable guide lanes. Although in recent years, the research on signal control optimization of variable lanes has gradually increased, it mainly focuses on the optimization of single type of variable lane, and the spatiotemporal coordination degree is low. In June 2017, the “Urban Road Traffic Civilization and Smoothness Promotion Action Plan (2017-2020)” jointly issued by the Ministry of Public Security and other departments clearly pointed out that according to the traffic flow variation law and road conditions, measures such as one-way traffic, variable lanes, tidal lanes, carpool lanes, etc. should be taken to alleviate the traffic congestion of main roads and important sections during peak hours.

[0003] In recent years, domestic and foreign scholars have included variable lane functions into the research scope of coordinated control optimization of urban trunk roads, mainly focusing on tidal lanes and variable guide lanes. In the aspect of tidal lanes, existing researches mainly focus on optimizing the layout scheme of tidal lanes, and the coordinated optimization of signal control is relatively less. Chinese patent CN206224805U discloses a variable lane control system for urban vehicle transportation carrying capacity, which can real-time judge the traffic demand of the road according to traffic flow data and weather data, and allocate the variable lane right on demand to alleviate the road traffic congestion. However, the current signal control optimization method of tidal lanes mainly adopts a step-by-step optimization strategy, i.e. determining the lane function first and then optimizing the signal control, which ignores the interactive influence between different control strategies and is easy to lead to local optimal solution. The signal timing optimization of variable guide lanes mainly focuses on the intersection level, and the influence of trunk level is less discussed. Although the potential of signal timing optimization of multiple types of variable lanes has been verified in the field application in many cities in China, it still mainly stays at the engineering practice level, and the layout rationality and timing optimization mainly depend on the experience of traffic managers. Further exploring the signal timing optimization of multiple types of variable lanes in theoretical research will significantly improve the traffic efficiency of the traffic system.

[0004] Therefore, the characteristics of different variable lanes should be analyzed, and a coordinated optimization model of urban trunk roads considering multiple types of variable lanes should be constructed to solve the traffic imbalance problem in urban traffic and increase the number of controllable objects in signal control. SUMMARY

[0005] The purpose of this invention is to overcome the shortcomings of the existing technology and provide a coordinated optimization method for urban arterial road signal control that considers multiple types of variable lanes, thereby achieving coordinated optimization of multiple types of variable lanes and signal control.

[0006] The objective of this invention can be achieved through the following technical solutions:

[0007] A method for coordinated optimization of signal control on urban arterial roads considering multiple types of variable lanes, the method comprising:

[0008] Based on the cumulative curve theory and collected data, cumulative arrival curves and cumulative departure curves of the outer sections of urban arterial roads are constructed. Based on the cumulative arrival curves and considering the differential distribution of driving behavior, virtual cumulative departure curves of the outer sections of urban arterial roads are constructed.

[0009] Based on the relationship between the cumulative arrival curve, the cumulative departure curve, and the virtual cumulative departure curve, the queuing distribution of the outer and inner sections of the urban arterial road is calculated respectively, and a multi-objective optimization function is constructed.

[0010] Considering vehicle conflict phenomena, number of lanes, traffic volume fluctuations, public cycle length of urban arterial roads, and traffic light duration, construct constraints for a multi-objective optimization function;

[0011] The multi-objective optimization function is solved using the particle swarm optimization algorithm to obtain the signal control scheme and lane division scheme, thereby realizing the coordinated optimization of urban arterial road signal control under multiple variable lane conditions.

[0012] Furthermore, the process of constructing the cumulative arrival curve and cumulative departure curve of the outer section of the urban arterial road includes: installing detectors on the outer section of the urban arterial road to be optimized, collecting traffic flow data, vehicle travel time data and traffic light data through the detectors, and constructing the cumulative arrival curve and cumulative departure curve based on the traffic flow data, vehicle travel time data, traffic light data and cumulative curve theory of the outer section of the urban arterial road to be optimized.

[0013] The cumulative arrival curve is constructed based on traffic data collected from the periphery of urban arterial roads and according to cumulative curve theory. The expression of the cumulative arrival curve is as follows:

[0014]

[0015] 0≤t≤E(TT M )+c

[0016] Among them, A M (t) represents the cumulative arrival curve of the flow from the outer section to M at time t, λ M The arrival rate of the flow from the outer section to M is represented by c, where c is the common period duration of the coordination range, and TT is the total flow rate. MLet M be the expected value of the free-flow travel time to the road segment where M is located;

[0017] The expression for the cumulative departure curve is:

[0018]

[0019] Among them, D M (t) represents the cumulative departure curve of the outer section towards M at time t, u M The green light is turned on for the current flow, v M The end time of the green light for the current flow, l M s represents the number of lanes currently flowing in the direction of the outer section M, i.e., the number of lanes available in that direction. M The vehicle dissipation rate in the current direction of the flow from the outer section M to the single lane is the number of vehicles that can pass through the single lane per unit time.

[0020] Furthermore, the process of constructing the virtual cumulative departure curve includes:

[0021] Considering the differences in driving behavior, the free-flow travel time of vehicles is assumed to follow a log-normal distribution;

[0022] Based on the cumulative arrival curve, calculate the joint distribution of the probability of vehicles arriving downstream at each time point;

[0023] Based on the joint distribution, the average number of vehicles arriving at the downstream section from the outer section at any given time is calculated, and a virtual cumulative departure curve is constructed.

[0024] The joint distribution of the probability of a vehicle arriving downstream at each time point is calculated as follows:

[0025]

[0026] in, This indicates that at time t, the cumulative number of arriving vehicles is N. M When (t), the probability that the number of vehicles arriving downstream is n is N. M (t) represents the cumulative number of vehicles arriving at M at a given time t, where t i V represents the timestamp of vehicle i passing through the upstream virtual section, V represents the sample space, that is, the different combinations of queued and non-queued vehicles in the current period, QV represents queued vehicles, and NQV represents non-queued vehicles.

[0027] N M (t)=λ M t

[0028]

[0029] Where, λ MThe arrival rate of the flow direction M at the outer section is represented by δ, which is a parameter characterizing the difference in driving behavior.

[0030] The average number of vehicles (PD) arriving at the downstream section from the outer section at any given time. M (t), that is, the calculation expression for the virtual cumulative departure curve, is:

[0031]

[0032] Furthermore, the queue distribution of the outer section of the urban arterial road includes the mean curve of the queue distribution, the upper bound curve of the queue distribution, and the lower bound curve of the queue distribution.

[0033] The expression for the mean curve of the queue distribution is:

[0034]

[0035] Where, q e (t) represents the mean curve of the queue distribution, k is the correction coefficient, and N M (t) represents the cumulative number of arriving vehicles at a given time t. Let N represent the cumulative number of arriving vehicles at time t. M When (t), the probability that the number of vehicles arriving downstream is n is given by D. M (t) represents the cumulative departure curve of the outer section towards M at time t;

[0036] The expression for the upper bound curve of the queue distribution is:

[0037]

[0038] Where, q u (t) is the upper bound curve of the queue distribution. This represents the sample size corresponding to the upper limit of the number of vehicles in the queue, that is, the statistical sample size corresponding to the maximum possible number of vehicles in the queue. P represents the sample size corresponding to the average number of vehicles in a queue, i.e., the statistical sample size corresponding to the average queue length. t,i This represents the probability that the number of vehicles in the queue is i at time t;

[0039] The expression for the lower bound curve of the queue distribution is:

[0040]

[0041] Where, q l (t) is the lower bound curve of the queue distribution.

[0042] Furthermore, the queue distribution within the cross-section of the urban arterial road includes the mean curve of the queue distribution, the upper bound curve of the queue distribution, and the lower bound curve of the queue distribution.

[0043] The calculation process for the queue distribution within the cross-section of the urban arterial road includes:

[0044] Model different internal sections of the current flow direction of the urban arterial road and calculate the cumulative departure curve of the upstream flow direction of different internal sections;

[0045] The time axis of each cumulative departure curve is shifted forward so that the time in the curve represents the moment within the upstream green light period;

[0046] Based on the mapping of the upstream green light period to the downstream cycle, the cumulative number of arriving vehicles in the downstream direction for different internal sections is calculated for different scenarios.

[0047] The cumulative number of arriving vehicles in the downstream direction of different internal sections is accumulated on the same time scale to obtain the total cumulative arrival curve of the internal sections;

[0048] Considering the differences in driving behavior, the free-flow travel time of vehicles is assumed to follow a log-normal distribution;

[0049] Based on the total cumulative arrival curve, calculate the joint distribution of the probability of vehicles arriving downstream at each time point;

[0050] Based on the joint distribution, the mean number of vehicles arriving at the downstream section from the upstream at any given time is calculated, and the total virtual cumulative departure curve of the internal section is constructed.

[0051] Based on the collected data on the green light start time, green light end time, number of lanes, and single-lane dissipation rate of the downstream flow direction, the total cumulative departure curve of the internal section is calculated and obtained.

[0052] Based on the total cumulative departure curve, total cumulative arrival curve, and total virtual cumulative departure curve of the internal cross-section of the urban arterial road, the queue distribution of the internal cross-section of the urban arterial road is calculated.

[0053] Furthermore, the expression for the cumulative departure curves of the upstream flow direction at different internal cross-sections is as follows:

[0054] AD M (t)=min[PD M (t),D M (t)]

[0055] Among them, PD M (t) represents the average number of vehicles arriving from upstream and flowing downstream to M at any given time, i.e., the virtual cumulative departure curve, D M (t) represents the cumulative departure curve from the upstream, AD M (t) represents the cumulative number of vehicles leaving upstream at time t;

[0056] The expression for the total cumulative arrival curve of the internal cross-section is:

[0057]

[0058] in, This represents the cumulative number of arriving vehicles provided by the upstream left-turning flow at time t. This represents the cumulative number of arriving vehicles provided by the upstream straight-through flow at time t. γ represents the cumulative number of arriving vehicles provided by the upstream right-turning flow at time t. N This indicates the steering ratio at which an upstream vehicle reaches the downstream.

[0059] The expression for the total virtual cumulative departure curve of the internal cross-section is:

[0060]

[0061] Among them, PD N (t) represents the average number of vehicles flowing towards N at time t downstream section. This indicates that at time t, the cumulative number of arriving vehicles is A. N The probability that the number of vehicles arriving downstream is n when (t) is reached.

[0062] The expression for the total cumulative departure curve of the internal cross-section is:

[0063]

[0064] Among them, D N (t) represents the cumulative departure curve of the internal cross-section towards N at time t, u N The green light is turned on for the current flow. N The end time of the green light flowing to N, l N s represents the number of lanes currently flowing in the direction N of the internal cross-section, i.e., the number of lanes available in that direction. N Let TT be the vehicle dissipation rate of the current flow into a single lane within the internal cross-section N, i.e., the number of vehicles that can pass through a single lane per unit time. N Let be the expected free-flow travel time to the road segment where N is located.

[0065] Furthermore, when calculating the cumulative number of arriving vehicles in the downstream direction for different internal cross-sections under different scenarios, the scenarios include:

[0066] The first scenario: the upstream green light period is mapped to the arrival of all downstream items within the downstream cycle;

[0067] The second scenario: the cumulative dissipation curve during the upstream green light period spans two cycles downstream;

[0068] In the first type of scenario, the formula for calculating the cumulative number of vehicles arriving downstream in different internal cross-sections is as follows:

[0069]

[0070] Among them, A M-N (t) represents the cumulative number of arriving vehicles at time t, x represents the offset of the dissipation curve of upstream flow M during the green light period relative to the time axis of downstream flow N, and r N This indicates the duration of the red light for the downstream flow direction being analyzed. This represents the phase difference between the downstream flow N and the upstream flow M, and the effective green light duration for the upstream flow M, TT. M Let M be the expected value of the free-flow travel time to the road segment where M is located;

[0071] In the second scenario, the formula for calculating the cumulative number of arriving vehicles in the downstream direction at different internal cross-sections is as follows:

[0072]

[0073] Among them, v N This indicates the end time of the green light for downstream flow to N.

[0074] Furthermore, the optimization objectives of the multi-objective optimization function include:

[0075] Category I objective: Traffic supply exceeds traffic demand, minimizing the maximum queue length within a cycle and reducing the phenomenon of idle green lights;

[0076] Category II objective: Traffic demand exceeds traffic supply, minimizing the maximum queue length within the cycle and alleviating oversaturation;

[0077] Category III objective: Supply exceeds demand for some items, demand exceeds supply for others, while balancing the green light vacancy and oversaturation;

[0078] The objective function for the first type of objective is:

[0079]

[0080] Where C represents the cycle length, M is the set of flow directions within the range to be optimized, and w M The weights for the flow to M, The upper bound curve for the queue distribution of the outer or inner sections of urban arterial roads;

[0081] The objective function for the second type of objective is:

[0082]

[0083] in, The lower bound curve for the queue distribution of the outer or inner sections of urban arterial roads;

[0084] The objective function for the third type of objective is:

[0085]

[0086] Where α is the weight between the Class I objective and the Class II objective.

[0087] Furthermore, the vehicle conflict phenomenon includes potential conflict phenomena of different flow directions at upstream and downstream intersections in the road segment and phase sequence conflict phenomena.

[0088] The potential conflicts between different flow directions at the upstream and downstream intersections on the road segment include:

[0089] Flow direction merging: Vehicles from two upstream directions converge into the same downstream direction;

[0090] Flow direction diversion: Vehicles flowing in the same upstream direction are diverted to different downstream directions;

[0091] Flow pair conflict: There is a vehicle weaving phenomenon between two flow pairs;

[0092] The constraints considering potential conflicts between upstream and downstream traffic flows at intersections within a road segment are as follows:

[0093]

[0094] Among them, u MP and v MP These represent the start and end times of the green light for the first flow direction; u MS and v MS E(·) and σ represent the start and end times of the green light for the subsequent traffic flow, respectively. 2 (·) represent the mean and variance of the free-flow travel time distribution in the direction of flow, respectively. This represents the mean of the free-flow travel time distribution towards MP, i.e., the average time it takes for a vehicle to travel through a road segment in free-flow condition. This represents the variance of the free-flow travel time distribution towards MP, i.e., the degree of fluctuation in free-flow travel time.

[0095] The constraints considering phase sequence conflict are as follows:

[0096]

[0097] Among them, l ij This represents the dissipation time between light group i and light group j, i.e., the safe interval before the next light group begins to allow vehicles to pass after the previous light group has completely passed through the intersection. jiu represents the dissipation time between lamp group j and lamp group i. i and v i Indicates the start and end times of the green light for light group i, u j and v j This indicates the start and end times of the green light for light group j, where c is the length of the common cycle, and w... ij Indicates the order in which conflict light groups i and j are released, a ij This indicates the conflict relationship between lamp group i and lamp group j, l k Let l be the time loss of lamp group k. h The time loss of lamp group h, u k and v k G represents the green light start and end times for light group k, respectively. k For the effective green light duration of light group k, g h The effective green light duration for light group h, l hk The dissipation time between lamp group h and lamp group k is represented by l. kh a represents the dissipation time between lamp group k and lamp group h. kh This indicates the conflict relationship between lamp group k and lamp group h;

[0098] The constraint considering the number of lanes is as follows:

[0099]

[0100] Among them, l M and l N These represent the number of lanes corresponding to flows to M and N, respectively. O and M I S represents the set of flow directions at the outer and inner cross sections within the urban arterial road area, respectively. M and S N Let {l, th, r} represent the flow direction combinations at the cross-sections where flow directions M and N are located, where l represents a left turn, th represents going straight, and r represents a right turn. and These represent the total number of lanes flowing towards sections M and N, respectively;

[0101] The constraints considering traffic volume fluctuations are as follows:

[0102] g k ≥fcb k

[0103] b k =q k / s k

[0104] Among them, g k b is the green light duration for the flow to k. k q represents the saturation level of the flow towards k.k With s k These represent the flow rate and capacity to k, respectively. f is a parameter that takes into account the range of traffic volume fluctuations. The larger the value, the greater the fluctuations that can be received.

[0105] The constraint considering the common cycle length of urban arterial roads is as follows:

[0106] c min ≤c≤c max

[0107] Where c is the common cycle length, which is the uniform cycle duration of signals at all intersections on the urban arterial road. min The minimum value representing the cycle length must meet the basic requirements for pedestrians crossing the street and vehicles starting up. c max This represents the maximum value of the period length;

[0108] The constraints considering the duration of traffic lights are as follows:

[0109]

[0110] Where k represents any group of lights, r k r represents the red light duration of light group k, that is, the duration for which the traffic light is red. kmin r represents the minimum red light duration for light group k. kmax g represents the maximum red light duration for light group k. kmin This represents the minimum effective green light duration for light group k, which must meet the basic needs of pedestrians crossing the street and vehicles passing through.

[0111] Furthermore, the process of solving the multi-objective optimization function based on the particle swarm optimization algorithm includes:

[0112] Particle sampling: Initialize the number of particles, parameters, and iteration termination conditions required for the particle swarm optimization algorithm;

[0113] The position of each particle is initialized based on the number of particles, where the position of each particle represents a set of variable values, i.e., an optimization scheme;

[0114] Fitness calculation: Based on the multi-objective optimization function, calculate the fitness value of each particle. The smaller the value of the optimization objective function, the higher the fitness of the corresponding particle.

[0115] Iterative decision: Calculate the historical best position of each particle, the historical best position of the population, and their fitness values, and update the velocity and position of the particles. Determine whether to end the calculation based on the iteration termination condition. If the termination condition is met, the calculation ends; otherwise, return to the particle sampling step.

[0116] Based on the final particle position and the constraints of the multi-objective optimization function, the optimization scheme within the urban arterial road area is reconstructed, including the signal control scheme and the lane division scheme, and then output.

[0117] Compared with the prior art, the beneficial effects of the present invention include:

[0118] 1. This invention addresses the problems of low coordination between existing urban arterial road signal control and variable lane optimization, limited scenario coverage, and a small number of controllable objects. It discloses a coordinated optimization method for urban arterial road signal control considering multiple types of variable lanes. By constructing cumulative arrival curves, cumulative departure curves, and virtual cumulative departure curves, a multi-objective optimization function and its constraints are established. The signal control scheme and lane division scheme are then obtained through solving this function, enabling coordinated optimization of multiple types of variable lanes and signal control. This invention improves upon the traditional cumulative curve diagram by constructing cumulative arrival and cumulative departure curves separately. Furthermore, considering the differential distribution of driving behavior, a virtual cumulative departure curve is innovatively constructed. The relationship between the three curves can be accurately... This invention accurately quantifies three scenarios on arterial roads: traffic supply exceeding demand (idle green lights), traffic demand exceeding supply (traffic congestion), and mixed traffic imbalance, significantly improving adaptability to complex traffic conditions. Although this invention is designed for urban arterial roads, the constructed signal control optimization function has strong applicability. It can be adapted and improved according to actual application scenarios to extend to various traffic scenarios such as single-point intersections and urban road networks, demonstrating high scalability. This invention has low requirements for input data, requiring only the arrival rate of each peripheral section as input, and the input can be obtained through various traffic detection methods. It is applicable to continuous spatiotemporal observation data and can also be extended to other traditional data sources that provide arrival rates, demonstrating strong adaptability.

[0119] 2. This invention innovatively constructs a virtual cumulative departure curve and estimates the mean, upper bound, and lower bound curves of the queue distribution based on the log-normal distribution of vehicle free-flow travel time, rather than the fixed queue length of traditional methods. This can accurately reflect the queue fluctuations caused by overtaking and other behaviors, providing quantitative basis for worst-case and best-case scenarios for subsequent optimization and improving the robustness of the model.

[0120] 3. This invention clearly divides the optimization objectives into three categories. Through the minimization and maximization logic of the objective function, it realizes the mathematical simulation of different objectives within the same framework, breaking the limitations of the traditional model of a single objective and a single scenario, and giving full play to the role of variable lanes in regulating spatiotemporal resources.

[0121] 4. This invention divides the main road into an outer section and an inner section. Based on the outer section, it establishes the cumulative arrival curve, cumulative departure curve and virtual cumulative departure curve of the inner section. When calculating the cumulative departure curve of the upstream flow direction of different inner sections, the upstream of the inner section is actually regarded as the outer section, which simplifies the calculation process.

[0122] 5. In this invention, when calculating the cumulative number of vehicles arriving in the downstream direction of different internal cross sections in different scenarios, the scenarios are divided into two categories: non-cross-cycle and cross-cycle. By offsetting the time axis and superimposing the cross-cycle flow, the arrival time deviation caused by the upstream and downstream phase difference is corrected, which is more in line with the actual traffic flow pattern than the traditional uniform arrival assumption.

[0123] 6. When calculating the total cumulative arrival curve of the internal section, this invention considers that the traffic flow of the internal section may come from multiple directions such as left turn, straight and right turn from upstream. By superimposing the arrival curves from multiple sources by steering ratio, the full-dimensional quantification of the queue distribution of the internal section is realized.

[0124] 7. This invention takes into account various vehicle conflict phenomena, number of lanes, traffic volume fluctuations, public cycle length of urban arterial roads, and traffic light duration, constrains the multi-objective optimization function, and ensures the safety and rationality of the optimization results. Attached Figure Description

[0125] Figure 1 This is a flowchart of the method of the present invention;

[0126] Figure 2 This is a schematic diagram of the vehicle accumulation curve in an embodiment of the present invention;

[0127] Figure 3 This is a schematic diagram of the queue distribution in an embodiment of the present invention;

[0128] Figure 4 This is a schematic diagram illustrating that demand exceeds supply in an embodiment of the present invention;

[0129] Figure 5 This is a schematic diagram illustrating a situation where supply exceeds demand in an embodiment of the present invention;

[0130] Figure 6 This is a schematic diagram illustrating a scenario where supply exceeds demand in an embodiment of the present invention;

[0131] Figure 7 This is a schematic diagram illustrating a scenario where demand exceeds supply in an embodiment of the present invention;

[0132] Figure 8 This is a schematic diagram illustrating the extrapolation range of the flow direction at the outer cross section in an embodiment of the present invention;

[0133] Figure 9 This is a schematic diagram of the first type of scenario for extrapolating the flow direction of the internal cross-section in an embodiment of the present invention;

[0134] Figure 10 This is a schematic diagram of the second type of scenario for the internal cross-sectional flow direction deduction range in an embodiment of the present invention;

[0135] Figure 11 This is a schematic diagram of the third type of scenario for the internal cross-sectional flow direction deduction range in an embodiment of the present invention;

[0136] Figure 12 This is a schematic diagram of the fourth type of scenario for the internal cross-section flow direction extrapolation range in an embodiment of the present invention;

[0137] Figure 13 This is a schematic diagram of the flow direction merging in an embodiment of the present invention;

[0138] Figure 14 This is a schematic diagram of the flow direction splitting in an embodiment of the present invention;

[0139] Figure 15 This is a schematic diagram of the flow direction conflict in an embodiment of the present invention;

[0140] Figure 16 This is a simplified diagram of the verification scenario channelization design in an embodiment of the present invention;

[0141] Figure 17 This is a schematic diagram of the peripheral section numbering and variable lane setting in an embodiment of the present invention;

[0142] Figure 18 This is a schematic diagram of the first type of verification scenario in an embodiment of the present invention;

[0143] Figure 19 This is a schematic diagram of the second type of verification scenario in an embodiment of the present invention;

[0144] Figure 20 This is a schematic diagram of the third type of verification scenario in an embodiment of the present invention. Detailed Implementation

[0145] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort should fall within the scope of protection of the present invention.

[0146] Example 1

[0147] This embodiment discloses a coordinated optimization method for signal control on urban arterial roads considering multiple types of variable lanes. The method is as follows: Figure 1As shown, the method includes steps S1-S4. In this method, the outer section is actually the entrance and exit of the urban arterial road, and the inner section is the cross-section of the intermediate intersection / road segment of the urban arterial road. The urban arterial road is composed of the outer section and the inner section. The queue distribution of the outer section reflects the traffic status at the boundary of the arterial road, and the queue distribution of the inner section reflects the traffic status inside the arterial road. This method obtains the signal control scheme and lane division scheme of each arterial road through multi-objective optimization, optimizes the lanes and signal control, and realizes the optimization of the queue situation of the outer section and the inner section, thereby reducing the congestion caused by unreasonable planning of urban arterial roads.

[0148] In this method, upstream and downstream are relative concepts. Upstream refers to the vehicle's origin end, which is the section / intersection that the vehicle passes through first in the direction of traffic flow. Conversely, downstream refers to the vehicle's destination end, which is the section / intersection that the vehicle passes through last in the direction of traffic flow.

[0149] The specific steps of this method are described below:

[0150] Step S1: Based on the cumulative curve theory and collected data, construct the cumulative arrival curve and cumulative departure curve of the outer section of the urban arterial road. Based on the cumulative arrival curve and considering the differential distribution of driving behavior, construct the virtual cumulative departure curve of the outer section of the urban arterial road.

[0151] Step S2: Based on the relationship between the cumulative arrival curve, the cumulative departure curve, and the virtual cumulative departure curve, a multi-objective optimization function is constructed, and the queuing distribution of the outer and inner sections of the urban arterial road is calculated respectively, and the multi-objective optimization function is quantitatively represented.

[0152] Step S3: Considering vehicle conflict phenomena, number of lanes, traffic volume fluctuations, length of public cycle on urban arterial roads, and traffic light duration, construct the constraints of the multi-objective optimization function.

[0153] Step S4: Solve the multi-objective optimization function based on the particle swarm optimization algorithm to obtain the signal control scheme and lane division scheme, and realize the coordinated optimization of urban arterial road signal control under multiple variable lane conditions.

[0154] In step S1, the cumulative arrival curve represents traffic demand, the cumulative departure curve represents traffic supply, and the virtual cumulative departure curve describes departure under free-flow conditions. This scheme specifically improves upon the traditional cumulative curve diagram, combining the cumulative arrival curve, cumulative departure curve, and virtual cumulative departure curve to accurately depict the traffic supply and demand relationship and enhance the ability to express different supply and demand states. Figure 2As shown, the relationship between the cumulative arrival curve, cumulative departure curve, and virtual cumulative departure curve reflects the queue distribution within a cycle. Based on this research framework, this method can consider driving behaviors such as overtaking to estimate the distribution of queue lengths, and thereby determine the mean, upper bound, and lower bound of the queue distribution. Figure 3 As shown.

[0155] If we replace the aforementioned cumulative departure curve with a cumulative departure vehicle capacity curve, we can represent the traffic supply within a cycle. If the virtual cumulative departure curve intersects the actual cumulative departure curve and the difference is negative, it geometrically reflects unused spatiotemporal resources. Specifically, if the virtual cumulative departure curve is still higher than the actual cumulative departure curve at the end of the cycle, it indicates that queuing exists, and traffic demand exceeds supply. Figure 4 As shown; conversely, if the virtual cumulative departure curve intersects with the actual cumulative departure curve before the end of the cycle and results in a negative value, it indicates that there is a green light idling phenomenon, and traffic supply exceeds demand, such as... Figure 5 As shown.

[0156] In step S1, the process of constructing the cumulative arrival curve and cumulative departure curve of the outer section of the urban arterial road includes:

[0157] Install detectors on the outer cross-section of the urban arterial road to be optimized;

[0158] Traffic flow data, vehicle travel time data, and traffic light data are collected through detectors;

[0159] Based on traffic flow data, vehicle travel time data, traffic light data, and cumulative curve theory of the outer section of the urban arterial road to be optimized, cumulative arrival curves and cumulative departure curves are constructed.

[0160] Assuming the analyzed flow direction is M = {m+1, E, th} (hereinafter referred to as flow direction M), that is, the straight-through flow direction at the east entrance (section d) of intersection m+1, other flow directions (E and th) can also be modeled in a similar way. When the analyzed section is the outer section, the input flow rate directly comes from the data collected by the detector, that is, the arrival rate of the analyzed flow direction is... Therefore, within the analyzed time period, the cumulative number of arriving vehicles at time t, i.e., the cumulative arrival curve, is expressed as:

[0161]

[0162] 0≤t≤E(TT M )+c

[0163] Among them, A M (t) represents the cumulative arrival curve of the flow from the outer section to M at time t, λ MThe arrival rate of the flow from the outer section to M is represented by c, where c is the common period duration of the coordination range, and TT is the total flow rate. M Let M be the expected value of the free-flow travel time to the road segment where M is located;

[0164] Cumulative departure curve D M (t) and the time u when the green light turns on for that flow direction M The green light ends at v M And the expected free-flow travel time TT M Relevant, such as Figure 8 As shown, the expression for the cumulative departure curve is as follows:

[0165]

[0166] Among them, D M (t) represents the cumulative departure curve of the outer section towards M at time t, u M The green light is turned on for the current flow. M The end time of the green light for the current flow, l M s represents the number of lanes currently flowing in the direction of the outer section M, i.e., the number of lanes available in that direction. M The vehicle dissipation rate in the current direction of the flow from the outer section M to the single lane is the number of vehicles that can pass through the single lane per unit time.

[0167] Based on the cumulative arrival curve, the process of constructing the virtual cumulative departure curve is as follows:

[0168] Considering the differences in driving behavior, the free-flow travel time of a vehicle on a road segment should conform to a certain distribution rather than a fixed value. Therefore, whether a vehicle can reach the downstream within a specified time should be regarded as a probabilistic event.

[0169] Assume that the free-flow travel time of the vehicle follows a log-normal distribution, i.e. For the input vehicle i, its arrival time is t. i At any given time t, its virtual departure time t = t i +TT f,*g It also satisfies a log-normal distribution, that is Therefore, within the analyzed time period, each vehicle has a probability value for whether it can reach the downstream, and the value of the virtual cumulative departure curve of the downstream section at each moment should be the joint distribution of the arrival probabilities of all vehicles.

[0170] Therefore, based on the cumulative arrival curve, the joint distribution of the probability of vehicles arriving downstream at each time point is calculated;

[0171] Subsequently, based on the joint distribution, the mean number of vehicles arriving at the downstream section from the outer section at any given time is calculated, and a virtual cumulative departure curve is constructed.

[0172] The joint distribution of the probability of vehicles arriving downstream at each time point is calculated as follows:

[0173]

[0174] in, This indicates that at time t, the cumulative number of arriving vehicles is N. M When (t), the probability that the number of vehicles arriving downstream is n is N. M (t) represents the cumulative number of vehicles arriving at M at a given time t, where t i V represents the timestamp of vehicle i passing through the upstream virtual section, V represents the sample space, which is the different combinations of queued and non-queued vehicles in the current period, QV represents queued vehicles, and NQV represents non-queued vehicles.

[0175] In this method, given the differences in driver behavior, including potential overtaking, the traditional first-come, first-served principle is no longer applicable. Therefore, the number of schemes in the sample space V can be determined as follows:

[0176]

[0177] Where, λ M The arrival rate of the flow direction M at the outer section is represented by δ, which is a parameter characterizing the difference in driving behavior.

[0178] The average number of vehicles (PD) arriving at the downstream section from the outer section at any given time. M (t), that is, the calculation expression for the virtual cumulative departure curve, is:

[0179]

[0180] Cumulative arrival curve A of the outer section M (t) depends only on the input traffic volume within the urban arterial road area, and the input traffic volume is assumed to be uniform, while the virtual cumulative departure curve PD M (t) is related to the road segment length and driving behavior. Based on this, the cumulative departure curve D can be further quantified. M (t). When reconstructing the outer section, the period offset TT needs to be adjusted. M This is to ensure that all vehicles arriving within the period are included in the calculation. Therefore, in reconstructing the cumulative arrival curve A... M (t), Virtual cumulative departure curve PD M (t) and the cumulative departure curve D M When (t), the time range should be [0, TT]. M +c].

[0181] Step S2 can be further divided into the following steps:

[0182] Step S201: Based on the relationship between the cumulative arrival curve, the cumulative departure curve, and the virtual cumulative departure curve, construct a multi-objective optimization function;

[0183] Step S202: Establish a queuing distribution estimation framework, estimate the queuing distribution of the inner and outer sections of the urban arterial road respectively, and quantify the objective function value.

[0184] In step S201, the multi-objective optimization function actually aims to minimize the maximum queuing length and periodic supply-demand imbalance of each flow direction within the research scope.

[0185] Specifically, traditional urban arterial road signal control optimization methods typically focus on addressing the problem of traffic demand exceeding supply, minimizing objectives such as maximum queue length or average vehicle delay through signal control. However, this method also introduces variable lanes, aiming to fully utilize spatiotemporal resources by coordinating lane function and signal timing. Optimizing only scenarios where demand exceeds supply cannot effectively reflect the goal of redistributing spatiotemporal resources; this method can optimize problems under various conditions.

[0186] This method categorizes potential optimization objectives, i.e., the optimization objectives of multi-objective optimization functions, into three types:

[0187] The first type of objective is when traffic supply exceeds traffic demand, and the need is to reduce idle green lights; the second type of objective is when traffic demand exceeds traffic supply, and the need is to reduce oversaturation; the third type of objective is when supply exceeds demand for some items and demand exceeds supply for others, and the need is to balance idle green lights and oversaturation.

[0188] The expected value of the queue length and its upper and lower bounds can be obtained from the relationship between the cumulative arrival curve, cumulative departure curve, and virtual cumulative departure curve obtained in step S1. Therefore, in the first type of objective, the optimization objective is to minimize the maximum queue length within the cycle and reduce the phenomenon of empty green lights. It should be ensured that the maximum queue length reaches its minimum value below the upper bound of the queue distribution, and at the end of the cycle, the phenomenon of empty green lights should reach its minimum value below the lower bound of the queue distribution. Figure 6 As shown. In the Type II objective, the optimization objective is to minimize the maximum queue length within a cycle and alleviate oversaturation. Therefore, it is necessary to reduce the maximum queue length and the upper bound of the queue length at the end of the cycle as much as possible, such as... Figure 7 As shown.

[0189] Furthermore, the optimization objectives of Type I and Type II can be summarized in two aspects: First, minimize the maximum queue length within the cycle, i.e., achieve the minimum value on the upper bound curve of the queue distribution, which is a typical minimax problem. Second, at the end of the cycle, minimize the absolute value of the queue distribution, making the queue distribution as close to 0 as possible, i.e., supply and demand are as balanced as possible. In Type I scenario, i.e., when traffic supply exceeds demand, the optimization should aim to minimize the absolute value on the lower bound curve of the queue distribution to address possible green light idling situations; in Type II scenario, i.e., when traffic demand exceeds supply, the optimization should aim to minimize the upper bound curve of the queue distribution to address possible oversaturation.

[0190] For a specific flow direction (such as left turn or straight ahead) in a certain cross section of the analysis, assuming the flow direction is numbered M, its cumulative arrival curve is A. M (t), the virtual cumulative departure curve is PD M (t), the actual cumulative departure curve is D M (t). Therefore, the mean curve of the queue distribution can be obtained. upper bound curve of queue distribution And queue distribution lower bound curve

[0191] Therefore, the objective function for the type I objective is expressed as follows. This objective function aims to minimize the maximum queue length for each flow direction within the study scope, taking into account the worst-case scenario of queue length:

[0192]

[0193] Where C represents the cycle length, M is the set of flow directions within the range to be optimized, and w M The weights for the flow to M, The upper bound curve for the queue distribution of the outer or inner sections of urban arterial roads;

[0194] Similarly, the objective function for the second type of objective is expressed as follows. This objective function aims to calculate the upper bound curve of the queue distribution at the end of the cycle. And queue distribution lower bound curve The value of is taken, and the maximum value is taken to represent the worst-case scenario of supply and demand imbalance:

[0195]

[0196] in, The lower bound curve for the queue distribution of the outer or inner sections of urban arterial roads;

[0197] In the above formula, by adjusting... and Taking the absolute value and then the maximum value, this approach comprehensively considers both oversaturation and the phenomenon of idle green lights, i.e., the discrepancy between traffic supply and demand. Therefore, scenario III (a hybrid scenario of scenarios I and II) can also be optimized based on this formula. The objective function expression for scenario III is as follows:

[0198]

[0199] Where α is the weight between the Class I objective and the Class II objective.

[0200] Within the urban arterial road area, due to the differences in data and geometric characteristics between the outer and inner cross-sections, the queuing distribution for different flow directions varies. The estimation process will also be different. The following explanation will cover the calculation of the queue distribution of the outer and inner sections.

[0201] In step S202, the queue distribution of both the outer and inner sections of the urban arterial road includes the mean curve of the queue distribution, the upper bound curve of the queue distribution, and the lower bound curve of the queue distribution.

[0202] Step S202.1: Estimate the queue distribution of the outer cross-section of the urban arterial road:

[0203] The expression for the mean curve of the queue distribution is:

[0204]

[0205] Where, q e (t) represents the mean curve of the queue distribution, k is the correction coefficient, and N M (t) represents the cumulative number of arriving vehicles at a given time t. Let N represent the cumulative number of arriving vehicles at time t. M When (t), the probability that the number of vehicles arriving downstream is n is given by D. M (t) represents the cumulative departure curve of the outer section towards M at time t;

[0206] The expression for the upper bound curve of the queue distribution is:

[0207]

[0208] Where, q u (t) is the upper bound curve of the queue distribution. This represents the sample size corresponding to the upper limit of the number of vehicles in the queue, that is, the statistical sample size corresponding to the maximum possible number of vehicles in the queue. P represents the sample size corresponding to the average number of vehicles in a queue, i.e., the statistical sample size corresponding to the average queue length. t,i This represents the probability that the number of vehicles in the queue is i at time t;

[0209] The expression for the lower bound curve of the queue distribution is:

[0210]

[0211] Where, q l (t) is the lower bound curve of the queue distribution.

[0212] Step S202.2, estimate the queue distribution of the inner section of the urban arterial road:

[0213] Because the input flow rate is constant, the construction of the cumulative arrival curve for the outer section is relatively easy. However, the cumulative arrival curve for the inner section is more complex because it is not only related to the input flow rate, but also needs to consider the timing of the upstream and downstream intersections. The complexity of its queue distribution construction is mainly reflected in two aspects: (1) The cumulative arrival curve needs to be based on the actual cumulative departure curve of the upstream, which needs to be modeled by combining the virtual cumulative departure curve and the actual cumulative departure curve of the upstream; (2) The start and end times of the time axis are different. When the upstream cumulative departure curve is mapped to the downstream, a time offset related to the timing parameters will be generated.

[0214] Assuming the analyzed flow direction is the straight-through flow at the east entrance (section j) of intersection m, other internal sections can be modeled using the same method. Regarding the first question, for the flow direction of internal sections, the upstream flow direction provides the actual cumulative departure curve AD. M (t) can be calculated using the following formula:

[0215] AD M (t)=min[PD M (t),D M (t)]

[0216] Among them, PD M (t) represents the average number of vehicles arriving from upstream and flowing downstream to M at any given time, i.e., the virtual cumulative departure curve, D M (t) represents the cumulative departure curve from the upstream, AD M (t) represents the cumulative number of vehicles leaving upstream at time t.

[0217] The formula is based on the virtual cumulative number of departing vehicles and the actual cumulative number of departing vehicles in the upstream direction. It calculates the minimum value of these two cumulative curves at any given time to construct the actual cumulative number of departing vehicles in the upstream direction.

[0218] The virtual cumulative number of departing vehicles and the actual cumulative number of departing vehicles in the upstream flow direction are calculated using the same method as the outer section in step S1 above.

[0219] Regarding the second question, the virtual cumulative departure curve of the upstream flow towards M and the time of the cumulative departure curve need to be shifted forward as a whole (uM +TT M ), that is, the variable minus (u M +TT M In this way, t in the obtained curve represents the time within the upstream green light period. For example, t = 1 represents the 1st second within the upstream green light period. Therefore, the variable t in the upstream cumulative departure curve after offset takes a value from 0 to the green light duration g of that flow direction. M Therefore, AD M (t) passes through (u) M +TT M The offset upstream cumulative departure curve will be directly used to calculate the cumulative number of vehicles arriving in the downstream direction.

[0220] However, the downstream internal cross-sectional flow direction differs from the upstream flow direction in terms of the start and end times on the time axis, exhibiting a certain offset. Based on the influence of the upstream and downstream phase difference and the mapping of the upstream green light period within the downstream cycle, four categories can be identified: (1) the upstream mapping arrives during the downstream red light period; (2) the upstream mapping arrives across both the red and green light periods within the downstream cycle; (3) the upstream mapping arrives during the downstream green light period; (4) the upstream mapping arrives across the current downstream green light period and the next downstream red light period, such as... Figures 9 to 12 As shown.

[0221] The first three categories can be grouped into one category, while the fourth category belongs to another.

[0222] The first scenario: the upstream green light period is mapped to the arrival of all downstream items within the downstream cycle;

[0223] The second scenario: the cumulative dissipation curve during the upstream green light period spans two cycles downstream.

[0224] In the first scenario, it is only necessary to find the deviation between the starting points of the two flows and then offset the cumulative dissipation curve during the upstream green light period on the time axis to obtain the cumulative arrival curve of the downstream flow. In the second scenario, since the cumulative dissipation curve during the upstream green light period spans two cycles downstream and the scenario under study is periodic, the arrival curve needs to be corrected.

[0225] For the first type of scenario, the time span of the analysis is 0 ≤ t ≤ TT. N +c, in the first type of scenario, the time axis deviation x satisfies the following relationship:

[0226]

[0227] Where x represents the offset of the upstream flow dissipation curve during the green light period relative to the time axis of the downstream analysis, and r N This indicates the duration of the red light for the downstream flow direction being analyzed. This represents the phase difference between the downstream flow direction N and the upstream flow direction M being analyzed.

[0228] The formula for calculating the cumulative number of downstream arriving vehicles at different internal cross-sections is as follows:

[0229]

[0230] Among them, A M-N (t) represents the cumulative number of arriving vehicles at time t, x represents the offset of the dissipation curve of upstream flow M during the green light period relative to the time axis of downstream flow N, and r N This indicates the duration of the red light for the downstream flow direction being analyzed. This represents the phase difference between the downstream flow N and the upstream flow M, and the effective green light duration for the upstream flow M, TT. M Let M be the expected value of the free-flow travel time to the road segment where M is located;

[0231] For the second scenario, the time axis offset x is calculated first. Then, the cumulative departure curve of the upstream flow during the green light period is divided into two parts: one part represents the portion that can reach the downstream current cycle within the expected free-flow travel time, directly mapped to the downstream flow; the other part represents the portion that will reach the downstream in the next cycle within the expected free-flow travel time. This part of the curve (denoted by z) can be superimposed on the beginning segment of the projected time period based on its periodicity. The length of curve z can be expressed by the following formula:

[0232]

[0233] For the second scenario, the formula for calculating the cumulative number of arriving vehicles in the downstream direction at different internal cross-sections is as follows:

[0234]

[0235] Among them, v N This indicates the end time of the green light for downstream flow to N.

[0236] For any internal cross-section flow direction, its arrival traffic volume may originate from upstream left-turning, straight-through, and right-turning flows. Let M∈{M1,M2,M3} represent the upstream left-turning, straight-through, and right-turning flows, respectively. Based on the above steps, the cumulative arrival curves provided by the upstream left-turning flow can be calculated separately. Cumulative arrival curve provided by upstream straight flow direction and the cumulative arrival curve of the upstream right-turning flow direction Since the upstream right-turn flow is uncontrolled, it can be calculated directly based on the arrival rate. By summing these three cumulative arrival curves on the same time scale, we can obtain the cumulative arrival vehicle curve A for the flow direction from upstream to downstream internal cross-section. N (t):

[0237]

[0238] Where, γ N This indicates the steering ratio at which an upstream vehicle reaches the downstream.

[0239] The cumulative arrival curve A of the internal cross-section flow direction is obtained. N After (t), the virtual cumulative departure curve PD of the internal section can be calculated using the same steps as in step S1. N The cumulative departure curves of (t) and internal cross sections accurately depict the supply and demand relationship.

[0240] Specifically, the expression for the total virtual cumulative departure curve of the internal cross-section is:

[0241]

[0242] Among them, PD N (t) represents the average number of vehicles flowing towards N at time t downstream section. This indicates that at time t, the cumulative number of arriving vehicles is A. N The probability that the number of vehicles arriving downstream is n when (t) is reached.

[0243] The expression for the total cumulative departure curve of the internal cross-section is:

[0244]

[0245] Among them, D N (t) represents the cumulative departure curve of the internal cross-section towards N at time t, u N The green light is turned on for the current flow, v N The end time of the green light flowing to N, l N s represents the number of lanes currently flowing in the direction N of the internal cross-section, i.e., the number of lanes available in that direction. N Let TT be the vehicle dissipation rate of the current flow into a single lane within the internal cross-section N, i.e., the number of vehicles that can pass through a single lane per unit time. N Let be the expected free-flow travel time to the road segment where N is located.

[0246] The next step is the same as step S202.1. Based on the total cumulative departure curve, total cumulative arrival curve and total virtual cumulative departure curve of the internal section of the urban arterial road, the queuing distribution of the internal section of the urban arterial road is calculated.

[0247] By substituting the upper and lower bound curves of the queue distribution of the outer or inner cross sections into the multi-objective optimization function, the quantitative construction of the multi-objective optimization function can be completed.

[0248] Step S3 is further divided into steps S301-S304 for different constraints.

[0249] Vehicle conflict phenomena include potential conflicts between different directions at upstream and downstream intersections on the road segment, as well as phase sequence conflicts.

[0250] Step S301: Construct constraints that take into account the potential conflicts of each flow direction at the upstream and downstream intersections in the road segment, with the goal of reducing lane conflicts.

[0251] Potential conflicts between different flow directions at upstream and downstream intersections on a road segment include:

[0252] Flow convergence: Vehicles flowing in two upstream directions converge into the same downstream direction, such as... Figure 13 As shown;

[0253] Flow direction diversion: Vehicles flowing in the same upstream direction are diverted to different downstream directions, such as... Figure 14 As shown;

[0254] Flow pair conflict: There is a vehicle weaving phenomenon between two flow pairs, such as... Figure 15 As shown.

[0255] Among the three relationships, the flow direction has the most significant impact on the conflict. When optimizing the signal timing scheme, priority should be given to distinguishing potential conflict flow pairs in the time dimension.

[0256] In a road segment consisting of two adjacent intersections, the potential merging points, diverging points, and conflict points are shown in Table 1 below.

[0257] Table 1 Classification of Road Section Conflicts

[0258] UL-DL UL-DT UL-DR UT-DL UT-DT UT-DR UR-DL UR-DT UR-DR UL-DL N F F H N N F N N UL-DT - N F C H N C H N UL-DR - - N C C H C C H UT-DL - - - N F F H N N UT-DT - - - - N F C H N UT-DR - - - - - N C C H UR-DL - - - - - - N F F UR-DT - - - - - - - N F UR-DR - - - - - - - - N

[0259] Where N represents no conflict, H represents merging flow pairs, F represents diverging flow pairs, and C represents conflicting flow pairs. UL, UT, and UR represent upstream left-turn, straight-ahead, and right-turn flows, respectively; DL, DT, and DR represent downstream left-turn, straight-ahead, and right-turn flows, respectively. For example, UL-DL represents the upstream left-turn to downstream left-turn flow. As shown in the table, there are 9 potential conflict points in the road segment formed by two adjacent intersections. Considering the uncontrolled nature of right-turn flows, conflict points related to upstream right-turn flows are unavoidable. Therefore, the remaining conflict points for flow pairs are: upstream left-turn to downstream straight-ahead flow and upstream straight-ahead to downstream left-turn flow; upstream left-turn to downstream right-turn flow and upstream straight-ahead to downstream left-turn flow; and upstream left-turn to downstream right-turn flow and upstream straight-ahead to downstream straight-ahead flow.

[0260] Considering the upper and lower bounds of the arrival queue distribution, for any pair of potential road segment conflict points, assuming MP is the flow direction that is released first and MS is the flow direction that is released later, the following constraints can be established to consider the potential conflict phenomena of each flow direction at the upstream and downstream intersections in the road segment:

[0261]

[0262] Among them, u MP and v MP These represent the start and end times of the green light for the first flow direction; u MS and v MS E(·) and σ represent the start and end times of the green light for the subsequent traffic flow, respectively. 2 (·) represent the mean and variance of the free-flow travel time distribution in the direction of flow, respectively. This represents the mean of the free-flow travel time distribution towards MP, i.e., the average time it takes for a vehicle to travel through a road segment in free-flow condition. This represents the variance of the free-flow travel time distribution towards MP, i.e., the degree of fluctuation in free-flow travel time.

[0263] Step S302: Construct constraints that take into account phase sequence conflict phenomena.

[0264] This method employs a traffic light group-based optimization approach, constraining the phase sequence. First, this method focuses on the left-turn and straight-through flows at each approach lane. Within the studied urban arterial road area, each intersection contains 8 flows, and a Boolean-type traffic light group conflict matrix NIM is constructed accordingly. Matrix element a ij This indicates the conflict relationship between lamp group i and lamp group j. When a ij When a = 1, it indicates that there is a conflict between lamp group i and lamp group j, and they need to be distinguished in terms of time; when a ij When the value is 0, it indicates that there is no conflict.

[0265] Arranged in the order of North Left Turn, North Straight, East Left Turn, East Straight, South Left Turn, South Straight, West Left Turn, and West Straight, the resulting traffic light conflict matrix is ​​a symmetric matrix, where each row or column contains at least one non-zero element, i.e., ∑ i a ij A value ≥1 indicates that all light groups may conflict with other light groups, as shown in the matrix below:

[0266]

[0267] Assume n is the number of intersections and c is the length of the common cycle. w is a constant greater than the period length. ij The direction sequence of conflicting light groups i and j is given by w. ij =0 indicates that light group i is released before light group j, wij =1 indicates that light group i is released after light group j. u k ,v k These represent the start and end times of the green light for light group k, respectively. k The time lost for lamp group k; l ij g is the dissipation time between lamp group i and lamp group j; k Let w be the effective green light duration for light group k. For any conflicting light groups i and j, the constraints can be constructed based on the green light activation and deactivation times, the release order, the dissipation time, and the cycle length. Specifically, if light group h is the last light group to be released within the cycle, then all other conflicting light groups are released before it, i.e., w. hk It is always 1.

[0268] Therefore, for the two cases mentioned above, the phase sequence must satisfy the following constraints, that is, the constraints considering the phase sequence conflict phenomenon are:

[0269]

[0270] Step S303: Construct constraints that take into account the number of lanes.

[0271] In this method, both tidal flow lanes and variable guidance lanes can be optimized. Based on actual conditions, lane functions can be converted into corresponding constraints, i.e., constraints considering the number of lanes:

[0272]

[0273] Among them, l M and l N These represent the number of lanes corresponding to flows to M and N, respectively. O and M I S represents the set of flow directions at the outer and inner cross sections within the urban arterial road area, respectively. M and S N Let {l, th, r} represent the flow direction combinations at the cross-sections where flow directions M and N are located, where l represents a left turn, th represents going straight, and r represents a right turn. and These represent the total number of lanes flowing towards sections M and N, respectively.

[0274] The above formula first constrains the lane function allocation of the outer section, that is, the sum of the number of lanes in each direction must not exceed the total number of lanes in the entrance lane. Then it is extended to the inner section, considering the layout of tidal lanes and variable guidance lanes, requiring that the total number of lanes in all directions (a total of 6 directions) of the road segment must not exceed the total number of lanes of the road segment.

[0275] Step S304: Construct constraints that take into account traffic volume fluctuations, the length of the public cycle of urban arterial roads, and the duration of traffic lights.

[0276] The constraints that take into account traffic volume fluctuations are designed to prevent oversaturation when traffic volume changes; their expression is as follows:

[0277] g k ≥fcb k

[0278] b k =q k / s k

[0279] Among them, g k b is the green light duration for the flow to k. k q represents the saturation level of the flow towards k. k With s k These represent the flow rate and capacity to k, respectively. f is a parameter that takes into account the range of traffic volume fluctuations. The larger the value, the greater the fluctuations that can be received. The minimum value is 1, and in this embodiment, the value is 1.2.

[0280] The common period length of the urban arterial roads under study must meet specified boundary conditions, i.e., the constraints on the common period length of urban arterial roads are considered:

[0281] c min ≤c≤c max

[0282] Where c is the common cycle length, which is the uniform cycle duration of signals at all intersections on the urban arterial road. min The minimum value representing the cycle length must meet the basic requirements for pedestrians crossing the street and vehicles starting up. c max This represents the maximum value of the period length;

[0283] For each light group k, the red and green light durations must satisfy specific boundary conditions to ensure the minimum green light duration required for pedestrians to cross the street. Considering factors such as pedestrian waiting time during the signal scheme design phase, the constraints on the traffic light durations are as follows:

[0284]

[0285] Where k represents any group of lights, r k r represents the red light duration of light group k, that is, the duration for which the traffic light is red. kmin r represents the minimum red light duration for light group k. kmax g represents the maximum red light duration for light group k. kmin This represents the minimum effective green light duration for light group k, which must meet the basic needs of pedestrians crossing the street and vehicles passing through.

[0286] Step S4, the process of solving the multi-objective optimization function based on the particle swarm optimization algorithm, includes:

[0287] Step S401, Particle sampling: Initialize the number of particles, parameters (such as weights), and iteration termination conditions required for the particle swarm algorithm (e.g., satisfying a specified number of iterations or the particle position change being less than a set threshold).

[0288] Step S402: Initialize the position of each particle according to the number of particles, where the position of each particle represents a set of variable values, i.e., an optimization scheme;

[0289] Step S403, Fitness Calculation: Calculate the fitness value of each particle according to the multi-objective optimization function. The smaller the value of the optimization objective function, the higher the fitness of the corresponding particle.

[0290] Step S404, Iterative judgment: Calculate the historical best position of each particle, the historical best position of the population and its fitness value, and update the velocity and position of the particles. Determine whether to end the calculation based on the iteration termination condition. If the termination condition is met, end the calculation; otherwise, return to step S401.

[0291] Step S405: Based on the final particle position and the constraints of the multi-objective optimization function, reconstruct the optimization scheme within the urban arterial road area, including the signal control scheme and the lane division scheme, and output the results.

[0292] Example 2

[0293] This embodiment, based on Embodiment 1 above, discloses a specific implementation example of a coordinated optimization method for signal control on urban arterial roads that considers multiple types of variable lanes.

[0294] This embodiment uses microscopic simulation software to construct a simulation model using a main road in a certain city as an example, and verifies the control optimization benefits of this method.

[0295] The simulation model is based on a certain road in a certain city, and it covers three intersections from east to west: Road A-Road B, Road A-Road C, and Road A-Road D. Figure 16 The channelization information for the verification scenario is displayed. Based on this, this method calibrates the simulation model using a timing scheme for the evening rush hour (7:30-9:00 PM) in November 2017, with a speed limit of 55 km / h. The timing scheme covers the signal timings of all intersections, such as... Figure 16 As shown, the three intersections implemented coordinated control under the original timing scheme, with global phase differences of 0 seconds, 18 seconds, and 91 seconds from left to right.

[0296] In this embodiment, the outer sections within the urban arterial road area are numbered (1 to 8), such as... Figure 17As shown in the diagram. During the verification, the eastern entrance of Chaoyang East Road-Yingzhou Road and the western entrance of Chaoyang East Road-Yuzhou Road were designated as reversible lane areas for illustrative verification. In this section, a tidal flow lane was installed, and each of the two entrances had a reversible lane.

[0297] Next, based on the constructed simulation model, traffic demand scenarios at different levels will be built.

[0298] 1) Verification Scenario 1: Balanced Traffic Flow on Main Roads. In this scenario, the traffic volume on the main urban arterial roads (eastbound and westbound traffic flows) is roughly balanced. The hourly traffic flows for westbound and eastbound traffic are 650 vehicles / hour and 550 vehicles / hour, respectively. Figure 18 As shown in Table 2, the path flow matrix is ​​presented below. This scenario is used to verify the signal coordination optimization capability of the method under general conditions (no significant directional imbalance).

[0299] Table 2. Path Traffic Matrix for the First Type of Verification Scenario

[0300]

[0301] 2) Verification Scenario 2: Traffic Volume Imbalance Scenario. In this scenario, there is a severe traffic volume imbalance between the intersection of Road A and Road B and the section between Road A and Road C. The hourly flow rate from west to east is 1080 vehicles / hour, while the flow rate from east to west is 339 vehicles / hour. The traffic volume imbalance coefficient is 0.76, exceeding 2 / 3 of the set tidal flow lane threshold. Figure 19 As shown in Table 3, the path traffic matrix is ​​presented below. This scenario is used to test the optimization effect of the method under unbalanced direction conditions.

[0302] Table 3. Path traffic matrix for the second type of verification scenario.

[0303]

[0304] 3) Verification Scenario 3: Scenario with coexisting directional and flow imbalances. In this scenario, there is a directional imbalance between the intersection of Road A and Road B and the section between Road A and Road C. The westbound flow is 840 vehicles / hour, and the eastbound flow is 339 vehicles / hour, with a directional imbalance coefficient of 0.71, which is greater than 2 / 3 of the tidal flow lane threshold. Furthermore, there is a flow imbalance at the western entrance of the Road A-Road C intersection, with left-turn and straight-through flow being roughly equal, but the lane allocation is 1:2. Figure 20 As shown in Table 4, the path flow matrix is ​​presented. This scenario is used to verify the adaptability of the method in complex situations where direction and flow are uneven.

[0305] Table 4. Path Traffic Matrix for the Third Type of Verification Scenario

[0306]

[0307] Based on the three types of verification scenarios constructed, three indicators were selected: average delay (AD), total throughput (TT) within the main road area, and average times of stop (AS) to conduct a quantitative analysis of the overall benefits.

[0308] The results for the three types of indicators—average vehicle stopping delay, total throughput within the main road area, and average number of stops—are shown in Table 5 below. The values ​​in parentheses represent the percentage improvement compared to the original plan. For the AD and AS indicators, negative values ​​indicate a decrease in average vehicle stopping delay or average number of stops, resulting in improved efficiency; positive values ​​indicate a decrease in efficiency. For the TT indicator, positive values ​​indicate an increase in total throughput, resulting in improved efficiency; negative values ​​indicate a decrease in efficiency.

[0309] Table 5. Benefits of Controlling the Area of ​​Main Roads

[0310]

[0311] As shown in the table above, our proposed method demonstrates good verification accuracy in all three verification scenarios. Overall, in scenarios without directional imbalance, the optimization control efficiency of our method is comparable to that of the MULTIBAND and SYNCHRO models. However, in scenarios with directional or flow imbalance, our method outperforms the comparative models.

[0312] Example 3

[0313] Based on Embodiment 1, this embodiment provides an electronic device, including: one or more processors and a memory, wherein the memory stores one or more programs, the one or more programs including instructions for executing the aforementioned coordinated optimization method for urban arterial road signal control considering multiple types of variable lanes.

[0314] At the hardware level, the electronic device includes a processor, internal bus, network interface, memory, and non-volatile memory, and may also include other hardware required for business operations. The processor reads the corresponding computer program from the non-volatile memory into memory and then runs it to implement the aforementioned method for coordinated optimization of urban arterial road signal control considering multiple types of variable lanes. Of course, in addition to software implementation, this invention does not exclude other implementation methods, such as logic devices or a combination of hardware and software, etc. That is to say, the execution subject of the following processing flow is not limited to individual logic units, but can also be hardware or logic devices.

[0315] Memory may include non-persistent storage in computer-readable media, such as random access memory (RAM) and / or non-volatile memory, such as read-only memory (ROM) or flash RAM. Memory is an example of computer-readable media.

[0316] Computer-readable media includes both permanent and non-permanent, removable and non-removable media that can store information using any method or technology. Information can be computer-readable instructions, data structures, modules of programs, or other data. Examples of computer storage media include, but are not limited to, phase-change memory (PRAM), static random access memory (SRAM), dynamic random access memory (DRAM), other types of random access memory (RAM), read-only memory (ROM), electrically erasable programmable read-only memory (EEPROM), flash memory or other memory technologies, CD-ROM, digital versatile optical disc (DVD) or other optical storage, magnetic tape, disk storage or other magnetic storage devices, or any other non-transferable medium that can be used to store information accessible by a computing device. As defined herein, computer-readable media does not include transient computer-readable media, such as modulated data signals and carrier waves.

[0317] The above description is merely a specific embodiment of the present invention, but the scope of protection of the present invention is not limited thereto. Any person skilled in the art can easily conceive of various equivalent modifications or substitutions within the technical scope disclosed in the present invention, and these modifications or substitutions should all be covered within the scope of protection of the present invention. Therefore, the scope of protection of the present invention should be determined by the scope of the claims.

Claims

1. A method for coordinated optimization of signal control on urban arterial roads considering multiple types of variable lanes, characterized in that, The method includes: Based on the cumulative curve theory and collected data, cumulative arrival curves and cumulative departure curves of the outer sections of urban arterial roads are constructed. Based on the cumulative arrival curves and considering the differential distribution of driving behavior, virtual cumulative departure curves of the outer sections of urban arterial roads are constructed. Based on the relationship between the cumulative arrival curve, the cumulative departure curve, and the virtual cumulative departure curve, the queuing distribution of the outer and inner sections of the urban arterial road is calculated respectively, and a multi-objective optimization function is constructed. Considering vehicle conflict phenomena, number of lanes, traffic volume fluctuations, public cycle length of urban arterial roads, and traffic light duration, construct constraints for a multi-objective optimization function; The multi-objective optimization function is solved using the particle swarm optimization algorithm to obtain the signal control scheme and lane division scheme, thereby realizing the coordinated optimization of urban arterial road signal control under multiple variable lane conditions.

2. The method for coordinated optimization of urban arterial road signal control considering multiple types of variable lanes as described in claim 1, characterized in that, The process of constructing the cumulative arrival curve and cumulative departure curve of the outer section of the urban arterial road includes: installing detectors on the outer section of the urban arterial road to be optimized, collecting traffic flow data, vehicle travel time data and traffic light data through the detectors, and constructing the cumulative arrival curve and cumulative departure curve based on the traffic flow data, vehicle travel time data, traffic light data and cumulative curve theory of the outer section of the urban arterial road to be optimized. The cumulative arrival curve is constructed based on traffic data collected from the periphery of urban arterial roads and according to cumulative curve theory. The expression of the cumulative arrival curve is as follows: 0≤t≤E(TT M )+c Among them, A M (t) represents the cumulative arrival curve of the flow from the outer section to M at time t, λ M The arrival rate of the flow from the outer section to M is represented by c, where c is the common period duration of the coordination range, and TT is the total flow rate. M Let M be the expected value of the free-flow travel time to the road segment where M is located; The expression for the cumulative departure curve is: Among them, D M (t) represents the cumulative departure curve of the outer section towards M at time t, u M The green light is turned on for the current flow, v M The end time of the green light for the current flow, l M s represents the number of lanes currently flowing in the direction of the outer section M, i.e., the number of lanes available in that direction. M The vehicle dissipation rate in the current direction of the flow from the outer section M to the single lane is the number of vehicles that can pass through the single lane per unit time.

3. The method for coordinated optimization of urban arterial road signal control considering multiple types of variable lanes as described in claim 1, characterized in that, The process of constructing the virtual cumulative departure curve includes: Considering the differences in driving behavior, the free-flow travel time of vehicles is assumed to follow a log-normal distribution; Based on the cumulative arrival curve, calculate the joint distribution of the probability of vehicles arriving downstream at each time point; Based on the joint distribution, the average number of vehicles arriving at the downstream section from the outer section at any given time is calculated, and a virtual cumulative departure curve is constructed. The joint distribution of the probability of a vehicle arriving downstream at each time point is calculated as follows: in, This indicates that at time t, the cumulative number of arriving vehicles is N. M When (t), the probability that the number of vehicles arriving downstream is n is N. M (t) represents the cumulative number of vehicles arriving at M at a given time t, where t i V represents the timestamp of vehicle i passing through the upstream virtual section, V represents the sample space, that is, the different combinations of queued and non-queued vehicles in the current period, QV represents queued vehicles, and NQV represents non-queued vehicles. N M (t)=λ M t Where, λ M The arrival rate of the flow direction M at the outer section is represented by δ, which is a parameter characterizing the difference in driving behavior. The average number of vehicles (PD) arriving at the downstream section from the outer section at any given time. M (t), that is, the calculation expression of the virtual cumulative departure curve, is:

4. The method for coordinated optimization of urban arterial road signal control considering multiple types of variable lanes as described in claim 1, characterized in that, The queue distribution of the outer section of the urban arterial road includes the mean curve of the queue distribution, the upper bound curve of the queue distribution, and the lower bound curve of the queue distribution. The expression for the mean curve of the queue distribution is: Where, q e (t) represents the mean curve of the queue distribution, k is the correction coefficient, and N M (t) represents the cumulative number of arriving vehicles at a given time t. Let N represent the cumulative number of arriving vehicles at time t. M When (t), the probability that the number of vehicles arriving downstream is n is given by D. M (t) represents the cumulative departure curve of the outer section towards M at time t; The expression for the upper bound curve of the queue distribution is: Where, q u (t) is the upper bound curve of the queue distribution. This represents the sample size corresponding to the upper limit of the number of vehicles in the queue, that is, the statistical sample size corresponding to the maximum possible number of vehicles in the queue. P represents the sample size corresponding to the average number of vehicles in a queue, i.e., the statistical sample size corresponding to the average queue length. t,i This represents the probability that the number of vehicles in the queue is i at time t; The expression for the lower bound curve of the queue distribution is: Where, q l (t) is the lower bound curve of the queue distribution.

5. The method for coordinated optimization of urban arterial road signal control considering multiple types of variable lanes as described in claim 1, characterized in that, The queue distribution within the cross-section of the urban arterial road includes the mean curve of the queue distribution, the upper bound curve of the queue distribution, and the lower bound curve of the queue distribution. The calculation process for the queue distribution within the cross-section of the urban arterial road includes: Model different internal sections of the current flow direction of the urban arterial road and calculate the cumulative departure curve of the upstream flow direction of different internal sections; The time axis of each cumulative departure curve is shifted forward so that the time in the curve represents the moment within the upstream green light period; Based on the mapping of the upstream green light period to the downstream cycle, the cumulative number of arriving vehicles in the downstream direction for different internal sections is calculated for different scenarios. The cumulative number of arriving vehicles in the downstream direction of different internal sections is accumulated on the same time scale to obtain the total cumulative arrival curve of the internal sections; Considering the differences in driving behavior, the free-flow travel time of vehicles is assumed to follow a log-normal distribution; Based on the total cumulative arrival curve, calculate the joint distribution of the probability of vehicles arriving downstream at each time point; Based on the joint distribution, the mean number of vehicles arriving at the downstream section from the upstream at any given time is calculated, and the total virtual cumulative departure curve of the internal section is constructed. Based on the collected data on the green light start time, green light end time, number of lanes, and single-lane dissipation rate of the downstream flow direction, the total cumulative departure curve of the internal section is calculated and obtained. Based on the total cumulative departure curve, total cumulative arrival curve, and total virtual cumulative departure curve of the internal cross-section of the urban arterial road, the queue distribution of the internal cross-section of the urban arterial road is calculated.

6. The method for coordinated optimization of urban arterial road signal control considering multiple types of variable lanes as described in claim 5, characterized in that, The expression for the cumulative departure curves of the upstream flow direction at different internal cross-sections is as follows: AD M (t)=min[PD M (t),D M (t)] Among them, PD M (t) represents the average number of vehicles arriving from upstream and flowing downstream to M at any given time, i.e., the virtual cumulative departure curve, D M (t) represents the cumulative departure curve from the upstream, AD M (t) represents the cumulative number of vehicles leaving upstream at time t; The expression for the total cumulative arrival curve of the internal cross-section is: in, This represents the cumulative number of arriving vehicles provided by the upstream left-turning flow at time t. This represents the cumulative number of arriving vehicles provided by the upstream straight-through flow at time t. γ represents the cumulative number of arriving vehicles provided by the upstream right-turning flow at time t. N This indicates the steering ratio at which an upstream vehicle reaches the downstream. The expression for the total virtual cumulative departure curve of the internal cross-section is: Among them, PD N (t) represents the average number of vehicles flowing towards N at time t downstream section. This indicates that at time t, the cumulative number of arriving vehicles is A. N The probability that the number of vehicles arriving downstream is n when (t) is reached. The expression for the total cumulative departure curve of the internal cross-section is: Among them, D N (t) represents the cumulative departure curve of the internal cross-section towards N at time t, u N The green light is turned on for the current flow, v N The end time of the green light flowing to N, l N s represents the number of lanes currently flowing in the direction N of the internal cross-section, i.e., the number of lanes available in that direction. N Let TT be the vehicle dissipation rate of the current flow into a single lane within the internal cross-section N, i.e., the number of vehicles that can pass through a single lane per unit time. N Let be the expected free-flow travel time to the road segment where N is located.

7. The method for coordinated optimization of urban arterial road signal control considering multiple types of variable lanes as described in claim 5, characterized in that, When calculating the cumulative number of arriving vehicles in the downstream direction for different internal cross-sections under different scenarios, the scenarios include: The first scenario: the upstream green light period is mapped to the arrival of all downstream items within the downstream cycle; The second scenario: the cumulative dissipation curve during the upstream green light period spans two cycles downstream; In the first type of scenario, the formula for calculating the cumulative number of vehicles arriving downstream in different internal cross-sections is as follows: Among them, A M-N (t) represents the cumulative number of arriving vehicles at time t, x represents the offset of the dissipation curve of upstream flow M during the green light period relative to the time axis of downstream flow N, and r N This indicates the duration of the red light for the downstream flow direction being analyzed. This represents the phase difference between the downstream flow N and the upstream flow M, and the effective green light duration for the upstream flow M, TT. M Let M be the expected value of the free-flow travel time to the road segment where M is located; In the second scenario, the formula for calculating the cumulative number of arriving vehicles in the downstream direction at different internal cross-sections is as follows: Among them, v N This indicates the end time of the green light for downstream flow to N.

8. The method for coordinated optimization of urban arterial road signal control considering multiple types of variable lanes according to claim 1, characterized in that, The optimization objectives of the multi-objective optimization function include: Category I objective: Traffic supply exceeds traffic demand, minimizing the maximum queue length within a cycle and reducing the phenomenon of idle green lights; Category II objective: Traffic demand exceeds traffic supply, minimizing the maximum queue length within the cycle and alleviating oversaturation; Category III objective: Supply exceeds demand for some items, demand exceeds supply for others, while balancing the green light vacancy and oversaturation; The objective function for the first type of objective is: Where C represents the cycle length, M is the set of flow directions within the range to be optimized, and w M The weights for the flow to M, The upper bound curve for the queue distribution of the outer or inner sections of urban arterial roads; The objective function for the second type of objective is: in, The lower bound curve for the queue distribution of the outer or inner sections of urban arterial roads; The objective function for the third type of objective is: Where α is the weight between the Class I objective and the Class II objective.

9. The method for coordinated optimization of urban arterial road signal control considering multiple types of variable lanes according to claim 1, characterized in that, The vehicle conflict phenomena include potential conflicts between different directions of flow at upstream and downstream intersections and phase sequence conflicts within the road segment. The potential conflicts between different flow directions at the upstream and downstream intersections on the road segment include: Flow direction merging: Vehicles from two upstream directions converge into the same downstream direction; Flow direction diversion: Vehicles flowing in the same upstream direction are diverted to different downstream directions; Flow pair conflict: There is a vehicle weaving phenomenon between two flow pairs; The constraints considering potential conflicts between upstream and downstream traffic flows in the road segment are as follows: Among them, u MP and v MP These represent the start and end times of the green light for the first flow direction; u MS and v MS E(·) and σ represent the start and end times of the green light for the subsequent traffic flow, respectively. 2 (·) represent the mean and variance of the free-flow travel time distribution in the direction of flow, respectively. This represents the mean of the free-flow travel time distribution towards MP, i.e., the average time it takes for a vehicle to travel through a road segment in free-flow condition. This represents the variance of the free-flow travel time distribution towards MP, i.e., the degree of fluctuation in free-flow travel time. The constraints considering phase sequence conflict are as follows: Among them, l ij This represents the dissipation time between light group i and light group j, i.e., the safe interval before the next light group begins to allow vehicles to pass after the previous light group has completely passed through the intersection. ji u represents the dissipation time between lamp group j and lamp group i. i and v i Indicates the start and end times of the green light for light group i, u j and v j This indicates the start and end times of the green light for light group j, where c is the length of the common cycle, and w... ij Indicates the order in which conflict light groups i and j are released, a ij This indicates the conflict relationship between lamp group i and lamp group j, l k Let l be the time loss of lamp group k. h The time loss of lamp group h, u k and v k G represents the green light start and end times for light group k, respectively. k For the effective green light duration of light group k, g h The effective green light duration for light group h, l hk The dissipation time between lamp group h and lamp group k is represented by l. kh a represents the dissipation time between lamp group k and lamp group h. kh This indicates the conflict relationship between lamp group k and lamp group h; The constraint considering the number of lanes is as follows: Among them, l M and l N These represent the number of lanes corresponding to flows to M and N, respectively. O and M I S represents the set of flow directions at the outer and inner cross sections within the urban arterial road area, respectively. M and S N Let {l, th, r} represent the flow direction combinations at the cross-sections where flow directions M and N are located, where l represents a left turn, th represents going straight, and r represents a right turn. and These represent the total number of lanes flowing towards sections M and N, respectively; The constraints considering traffic volume fluctuations are as follows: g k ≥fcb k b k =q k / s k Among them, g k b is the green light duration for the flow to k. k q represents the saturation level of the flow towards k. k With s k These represent the flow rate and capacity to k, respectively. f is a parameter that takes into account the range of traffic volume fluctuations. The larger the value, the greater the fluctuations that can be received. The constraint considering the common cycle length of urban arterial roads is as follows: c min ≤c≤c max Where c is the common cycle length, which is the uniform cycle duration of signals at all intersections on the urban arterial road. min The minimum value representing the cycle length must meet the basic requirements for pedestrians crossing the street and vehicles starting up. c max This represents the maximum value of the period length; The constraints considering the duration of traffic lights are as follows: Where k represents any group of lights, r k r represents the red light duration of light group k, that is, the duration for which the traffic light is red. kmin r represents the minimum red light duration for light group k. kmax g represents the maximum red light duration for light group k. kmin This represents the minimum effective green light duration for light group k, which must meet the basic needs of pedestrians crossing the street and vehicles passing through.

10. The method for coordinated optimization of urban arterial road signal control considering multiple types of variable lanes according to claim 1, characterized in that, The process of solving the multi-objective optimization function based on the particle swarm optimization algorithm includes: Particle sampling: Initialize the number of particles, parameters, and iteration termination conditions required for the particle swarm optimization algorithm; The position of each particle is initialized based on the number of particles, where the position of each particle represents a set of variable values, i.e., an optimization scheme; Fitness calculation: Based on the multi-objective optimization function, calculate the fitness value of each particle. The smaller the value of the optimization objective function, the higher the fitness of the corresponding particle. Iterative decision: Calculate the historical best position of each particle, the historical best position of the population, and their fitness values, and update the velocity and position of the particles. Determine whether to end the calculation based on the iteration termination condition. If the termination condition is met, the calculation ends; otherwise, return to the particle sampling step. Based on the final particle position and the constraints of the multi-objective optimization function, the optimization scheme within the urban arterial road area is reconstructed, including the signal control scheme and the lane division scheme, and then output.

Images