A queuing overflow control method and system in a second-level adaptive control system

Abstract

A queuing overflow control method and system in a second-level adaptive control system, which relates to the field of electronic digital data processing. The method includes: collecting real-time traffic data of each approach lane, calculating the queuing ratio and vehicle accumulation rate of each approach lane according to the real-time traffic data, and dividing the traffic state of each approach lane into a normal state, a warning state, and a dangerous state according to a preset state threshold; when the traffic state is in the warning state, generating a traffic flow time series curve according to the real-time traffic data; extracting traffic flow characteristic parameters from the traffic flow time series curve, arranging the traffic flow characteristic parameters according to the approach lane numbers to construct a traffic flow characteristic matrix, analyzing the traffic flow complementary relationship between approach lanes based on the traffic flow characteristic matrix, and calculating phase compensation parameters according to the traffic flow complementary relationship; adjusting the signal timing sequence according to the phase compensation parameters. Implementing this method can improve the balance of resource utilization in traffic approach intersections.

Description

Technical Field

[0001] This application relates to the field of electronic digital data processing, and particularly to a queuing overflow control method and system in a second-level adaptive control system. Background Art

[0002] With the acceleration of the urbanization process and the continuous growth of the motor vehicle ownership, the problem of urban traffic congestion has become increasingly prominent. As an important means to alleviate traffic congestion, the control effect of the urban traffic signal control system directly affects the traffic efficiency of urban roads and the traffic operation quality.

[0003] In the related art, a fixed timing scheme or an adaptive control scheme based on historical data can be adopted. These control schemes adjust the vehicle passage at intersections through preset parameters such as cycle and green ratio, or perform periodic signal timing optimization based on the traffic flow data in the historical database, so as to realize the dredging of traffic flow.

[0004] However, due to the short-term sudden fluctuations in traffic flow, the signal timing scheme cannot respond to the instantaneous changes in traffic flow in a timely manner. At the same time, the dynamic change rules of the traffic flow between each approach lane are not fully utilized, resulting in the phenomenon that while there is a queuing overflow in some approach lanes, there is a waste of green light resources in other approach lanes. Summary of the Invention

[0005] This application provides a queuing overflow control method and system in a second-level adaptive control system, which is used to improve the balance of resource utilization in traffic approach lanes.

[0006] In a first aspect, this application provides a queuing overflow control method in a second-level adaptive control system, which is applied to a second-level adaptive control system. The method includes: collecting real-time traffic data of each approach lane, where the real-time traffic data includes vehicle queue length, vehicle arrival rate, and vehicle departure rate; calculating the queuing ratio and vehicle accumulation rate of each approach lane according to the real-time traffic data, and classifying the traffic state of each approach lane into a normal state, a warning state, and a dangerous state according to a preset state threshold; when the traffic state is in the warning state, generating a traffic flow time series curve according to the real-time traffic data; extracting traffic flow characteristic parameters from the traffic flow time series curve, where the traffic flow characteristic parameters include the maximum traffic flow, the minimum traffic flow, the occurrence time of the maximum value, and the occurrence time of the minimum value; arranging the traffic flow characteristic parameters according to the approach lane number to construct a traffic flow characteristic matrix; analyzing the complementary relationship of traffic flow between approach lanes based on the traffic flow characteristic matrix, and calculating a phase compensation parameter according to the complementary relationship of traffic flow; and adjusting the signal timing sequence according to the phase compensation parameter.

[0007] By adopting the above technical solution, traffic states are divided by collecting traffic data in real time and calculating the queuing ratio and vehicle accumulation rate. Under the warning state, a time series curve of vehicle flow is generated and characteristic parameters are extracted to construct a characteristic matrix. This characteristic matrix effectively captures the short-term sudden fluctuation characteristics of traffic flow, and by analyzing the complementary relationship of vehicle flow between approach lanes, phase compensation parameters are calculated to dynamically adjust signal timing, making full use of the dynamic change law of vehicle flow between each approach lane and avoiding the problem of coexistence of queuing overflow at some approach lanes and waste of green light resources at other approach lanes.

[0008] Combined with some embodiments of the first aspect, in some embodiments, the step of analyzing the complementary relationship of vehicle flow between approach lanes based on the vehicle flow characteristic matrix and calculating phase compensation parameters according to the complementary relationship of vehicle flow specifically includes: arranging the vehicle flow characteristic matrix in chronological order to form a time series matrix; calculating the vehicle flow correlation coefficient between any two approach lanes according to the time series matrix; determining the combination of approach lanes with a vehicle flow correlation coefficient less than a preset threshold as the combination of approach lanes with a complementary relationship; calculating the time interval between the moment of the maximum vehicle flow of the first approach lane and the moment of the minimum vehicle flow of the second approach lane in the combination of approach lanes with a complementary relationship; using this time interval as the initial value of the phase compensation parameter; and performing weighted correction on the initial value of the phase compensation parameter according to the ratio of the maximum vehicle flow of the first approach lane to the minimum vehicle flow of the second approach lane to obtain the phase compensation parameter.

[0009] By adopting the above technical solution, the vehicle flow characteristic matrix is arranged in chronological order to form a time series matrix, the vehicle flow correlation coefficient between approach lanes is calculated, and the combination of approach lanes with a complementary relationship is determined. Based on the time interval between the moment of the maximum vehicle flow and the moment of the minimum vehicle flow as the initial value, combined with the vehicle flow ratio for weighted correction, a more accurate phase compensation parameter is obtained, optimizing the timing adjustment accuracy of signal timing and making the traffic flow conversion smoother.

[0010] Combined with some embodiments of the first aspect, in some embodiments, the step of adjusting the signal timing sequence according to the phase compensation parameter specifically includes: subtracting the phase compensation parameter from the current phase difference to obtain the target phase difference; adjusting the difference between the phase start moment of the first approach lane and the phase start moment of the second approach lane in the combination of approach lanes with a complementary relationship to the target phase difference; setting the first green light phase duration according to the maximum vehicle flow of the first approach lane, and setting the second green light phase duration according to the ratio of the maximum vehicle flow to the minimum vehicle flow of the second approach lane; distributing the remaining duration to the green light phases of the third approach lane and the fourth approach lane according to the ratio of the maximum vehicle flow; and determining the red light phase duration of each approach lane according to the difference between the phase start moments, and the sum of the green light phase duration and the red light phase duration of each approach lane is equal to the total cycle duration.

[0011] By adopting the above technical solution, the starting moment of the phase of the complementary relationship approach is adjusted according to the target phase difference, the first green light phase duration is set based on the maximum traffic flow, and the second green light phase duration is set according to the ratio of the maximum traffic flow to the minimum traffic flow. The remaining duration is proportionally allocated to other approachs, realizing the dynamic optimization allocation of the green light phase duration and improving the scientificity and rationality of the signal timing plan.

[0012] Combined with some embodiments of the first aspect, in some embodiments, after the step of calculating the queuing ratio and vehicle accumulation rate of each approach according to the real-time traffic data and dividing the traffic state of each approach into a normal state, a warning state, and a dangerous state according to a preset state threshold, the method further includes: when the traffic state is in the normal state, obtaining the vehicle gap acceptance rate of the approach; when the vehicle gap acceptance rate is greater than a preset acceptance rate threshold, shortening the signal timing sequence by a preset time length; when the vehicle gap acceptance rate is not greater than the preset acceptance rate threshold, extending the signal timing sequence by a preset time length.

[0013] By adopting the above technical solution, the vehicle gap acceptance rate of the approach is obtained under normal conditions, and the length of the signal timing sequence is dynamically adjusted according to the comparison result with the preset acceptance rate threshold. When the acceptance rate is greater than the threshold, the timing sequence is shortened, and when it is less than the threshold, the timing sequence is extended, realizing the refined adaptive adjustment of signal timing, improving the traffic efficiency and reducing the vehicle waiting time.

[0014] Combined with some embodiments of the first aspect, in some embodiments, after the step of extending the signal timing sequence by a preset time length when the vehicle gap acceptance rate is not greater than the preset acceptance rate threshold, the method further includes: real-time monitoring the change trends of the queuing ratio and the vehicle accumulation rate; when it is detected that the queuing ratio continuously rises in multiple consecutive signal cycles and exceeds the first warning threshold, switching the traffic state to the warning state.

[0015] By adopting the above technical solution, the change trends of the queuing ratio and the vehicle accumulation rate are monitored in real time. When the queuing ratio continuously rises in multiple consecutive signal cycles and exceeds the first warning threshold, the traffic state is timely switched to the warning state. A dynamic state switching mechanism is established to accurately grasp the evolution law of the traffic state and ensure the timely responsiveness of the traffic control system.

[0016] In combination with some embodiments of the first aspect, in some embodiments, after the steps of calculating the queuing ratio and vehicle accumulation rate of each approach according to the real-time traffic data and classifying the traffic state of each approach into a normal state, a warning state, and a dangerous state according to a preset state threshold, the method further includes: when the traffic state is in the dangerous state, calculating the vehicle saturation of each approach, where the vehicle saturation is equal to the product of the vehicle arrival rate and the vehicle gap acceptance rate; sorting the approaches according to the vehicle saturation; and extending the green light time period of the approach with the highest priority to a preset time.

[0017] By adopting the above technical solution, the vehicle saturation of each approach is calculated in the dangerous state, the priority is sorted based on the product of the vehicle arrival rate and the gap acceptance rate, and the green light time period of the approach with the highest priority is extended to a preset time. A dynamic priority allocation mechanism based on saturation is constructed, effectively alleviating the traffic pressure on the approaches with a relatively high degree of traffic congestion.

[0018] In combination with some embodiments of the first aspect, in some embodiments, after the step of extending the green light time period of the approach with the highest priority to a preset time, the method further includes: when the queuing ratio drops below a second warning threshold and the vehicle accumulation rate shows a downward trend, and at the same time the vehicle gap acceptance rate is within the preset normal range, switching the traffic state to the warning state.

[0019] By adopting the above technical solution, the state where the queuing ratio drops below the second warning threshold and the vehicle accumulation rate shows a downward trend is monitored, and at the same time, in combination with whether the vehicle gap acceptance rate is within the normal range, the automatic switching of the traffic state from the dangerous state to the warning state is realized. A joint determination mechanism for multi-dimensional indicators is established, accurately identifying the inflection point of the improvement of the traffic condition, realizing the smooth transition of the traffic control state, and avoiding the fluctuation of the traffic efficiency caused by frequent switching of the control state.

[0020] In a second aspect, an embodiment of the present application provides a second-level adaptive control system, which includes: one or more processors and a memory; the memory is coupled to the one or more processors, and the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the second-level adaptive control system to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0021] In a third aspect, an embodiment of the present application provides a computer program product containing instructions, which, when the computer program product runs on the second-level adaptive control system, enables the second-level adaptive control system to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0022] Fourthly, an embodiment of the present application provides a computer-readable storage medium, including instructions, which, when running on the second-level adaptive control system, cause the second-level adaptive control system to execute the method described in the first aspect and any possible implementation manner in the first aspect.

[0023] It can be understood that the second-level adaptive control system provided in the second aspect, the computer program product provided in the third aspect, and the computer storage medium provided in the fourth aspect are all used to execute the method provided in the embodiments of the present application. Therefore, the beneficial effects that can be achieved can refer to the beneficial effects in the corresponding method, and will not be elaborated here.

[0024] One or more technical solutions provided in the embodiments of the present application have at least the following technical effects or advantages:

[0025] 1. In the present application, traffic data is collected in real time, the queuing ratio and vehicle accumulation rate are calculated for traffic state division, a traffic flow time series curve is generated in the warning state, and characteristic parameters are extracted to construct a characteristic matrix. This characteristic matrix effectively captures the short-term sudden fluctuation characteristics of traffic flow, and by analyzing the complementary relationship of traffic flow between approach lanes, the phase compensation parameter is calculated to dynamically adjust the signal timing, making full use of the dynamic change law of traffic flow between each approach lane, and avoiding the problem of coexistence of queuing overflow at some approach lanes and waste of green light resources at other approach lanes.

[0026] 2. In the present application, the traffic flow characteristic matrix is arranged in chronological order to form a time series matrix, the correlation coefficient of traffic flow between approach lanes is calculated, and the combination of complementary approach lanes is determined. Based on the time interval between the maximum traffic flow moment and the minimum traffic flow moment as the initial value, combined with the traffic flow ratio for weighted correction, a more accurate phase compensation parameter is obtained, optimizing the timing adjustment accuracy of signal timing and making the traffic flow conversion smoother.

[0027] 3. In the present application, the phase start time of the complementary approach lanes is adjusted according to the target phase difference, the first green light phase duration is set according to the maximum traffic flow, and the second green light phase duration is set according to the ratio of the maximum traffic flow to the minimum traffic flow. The remaining duration is proportionally allocated to other approach lanes, realizing the dynamic optimal allocation of the green light phase duration and improving the scientificity and rationality of the signal timing scheme. Description of the Drawings

[0028] Figure 1 is a flowchart of a queuing overflow control method in the second-level adaptive control system in the embodiments of the present application;

[0029] Figure 2 is another flowchart of a queuing overflow control method in the second-level adaptive control system in the embodiments of the present application;

[0030] Figure 3 It is a schematic structural diagram of an entity device in the second-level adaptive control system in the embodiments of the present application. Specific embodiments

[0031] The terms used in the following embodiments of the present application are only for the purpose of describing specific embodiments and are not intended to limit the present application. As used in the specification of the present application, the singular forms "a", "an", "the above", "the", and "this" are also intended to include the plural forms unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in the present application refers to any or all possible combinations including one or more of the listed items.

[0032] Hereinafter, the terms "first" and "second" are only used for descriptive purposes and cannot be construed as implying or suggesting relative importance or implicitly indicating the quantity of the indicated technical features. Thus, features defined with "first" and "second" may explicitly or implicitly include one or more of such features. In the description of the embodiments of the present application, unless otherwise stated, the meaning of "a plurality" is two or more.

[0033] For ease of understanding, the application scenarios of the embodiments of the present application are introduced below.

[0034] At a busy intersection, severe traffic congestion often occurs during peak hours. The main east-west road intersects with the secondary north-south road at this intersection, and the traffic flow distribution is extremely unbalanced during peak hours. During the morning peak, the traffic flow on the eastbound approach suddenly increases, and the queue length rapidly grows to 300 meters. At the same time, the traffic flow on the southbound approach is sparse, and a large amount of green light time is not fully utilized during the green light period. At noon, the situation reverses, the traffic flow on the southbound approach surges, while the traffic flow on the eastbound approach significantly decreases. This short-term and drastic fluctuation of the traffic flow makes it difficult for the fixed signal timing plan to respond in a timely manner, resulting in queue overflow at some approaches and waste of green light resources at other approaches. Especially when the queue length of a certain approach exceeds 300 meters, it will also cause the traffic operation of the upstream intersection to be blocked, triggering a chain reaction. In this case, there is an urgent need for an adaptive control method that can quickly respond to traffic flow mutations and make full use of the green light resources of each approach.

[0035] In the current traffic signal control system, usually two solutions are adopted to handle the above problems. The first one is to use a fixed timing plan, where the signal timing plans for the early, middle, and late periods are preset. For example, during the morning peak period, the green light duration for the eastbound approach is 60 seconds, and for the southbound approach is 30 seconds; during the noon period, the eastbound is 40 seconds, and the southbound is 50 seconds. However, this solution cannot cope with sudden traffic flow fluctuations outside the preset periods. The second one is an adaptive control plan based on historical data. The system counts the traffic flow data every 5 minutes, predicts the traffic flow trend for the next 5 minutes according to the historical data of the same period, and dynamically adjusts the signal timing. However, when there are sudden traffic flow fluctuations, such as a 200% sudden increase in the traffic volume of a certain approach within 1 minute due to an emergency, the 5-minute response cycle is significantly lagging, and the signal timing cannot be adjusted in time, resulting in queue overflow. At the same time, both of these solutions fail to make full use of the complementary characteristics of the traffic volume among approaches, resulting in low overall system efficiency.

[0036] After adopting the second-level adaptive control system of this application, the operation effect of this intersection has been significantly improved. The system collects traffic flow data every 20 seconds through video detectors to quickly perceive changes in traffic conditions. When there is a sudden increase in the traffic volume of the eastbound approach, the system immediately calculates that the queue ratio exceeds 0.6, triggering an early warning state. Subsequently, the system analyzes and finds that the traffic volume of the southbound approach is at a low ebb, and the correlation coefficient between the two approaches is -0.8, confirming an obvious complementary relationship. The system calculates the optimal phase compensation parameter of 15 seconds, advances the starting moment of the eastbound green light, and adjusts the eastbound green light duration to 55 seconds and the southbound to 35 seconds according to the traffic flow ratio. This second-level adjustment based on the complementary relationship of traffic flow enables the queuing vehicles in the east direction to quickly disperse within 2 - 3 cycles, avoiding queue overflow, and at the same time, the green light resources of the southbound approach are reasonably utilized. Practice shows that this solution can reduce the average delay at the intersection by 25% and improve the traffic efficiency by 30%.

[0037] For ease of understanding, the method provided in this embodiment will be described in a process below in combination with the above scenario. Please refer to Figure 1 , which is a schematic flow diagram of the queue overflow control method in the second-level adaptive control system in the embodiment of this application.

[0038] S101. Collect the real-time traffic data of each approach, where the real-time traffic data includes the vehicle queue length, vehicle arrival rate, and vehicle departure rate.

[0039] Among them, the approach road refers to the vehicle passage section connected to the intersection. Real-time traffic data refers to the data set reflecting the current traffic operation state obtained through detection devices. The vehicle queue length refers to the total length of the vehicle queue waiting to pass through the intersection, with the unit of meter. The vehicle arrival rate refers to the number of vehicles arriving at the intersection per unit time, with the unit of vehicle / second. The vehicle departure rate refers to the number of vehicles leaving the intersection per unit time, with the unit of vehicle / second. Detection devices include loop detectors, video detectors, radar detectors, etc. The traffic data collection period is usually set to 20 seconds.

[0040] This step starts to execute after the system starts and completes the initialization configuration, and is used to continuously obtain traffic operation state data. Specifically, the system first checks the working status of each detection device to confirm that the data collection channel is normal. Then, according to the preset sampling period, it synchronously collects the detection data of each approach road. For the vehicle queue length, it is obtained by video image analysis or loop occupancy time calculation. For the vehicle arrival rate, it is obtained by recording the number of passing vehicles by the upstream detector and dividing it by the time interval. For the vehicle departure rate, it is obtained by recording the number of passing vehicles by the downstream detector and dividing it by the time interval.

[0041] In some embodiments, traffic data collection and processing can be achieved in various ways: Optionally, use the video detection scheme: deploy high-definition cameras to collect vehicle images, use deep learning algorithms for vehicle recognition and tracking, extract vehicle position and speed information, calculate the queue length and flow parameters, and transmit the processing results to the controller; Optionally, use the loop detection scheme: bury induction loops on the road surface, collect the loop impedance change signals through detectors, calculate the vehicle occupancy time and passing time, and convert them into the queue length and flow data in combination with the pre-calibrated parameter model, and upload the data to the controller. It can be understood that other methods such as radar detection and ultrasonic detection can also be used to achieve traffic data collection, which is not limited here.

[0042] S102. Calculate the queue ratio and vehicle accumulation rate of each approach road according to the real-time traffic data, and divide the traffic state of each approach road into a normal state, a warning state, and a dangerous state according to the preset state threshold.

[0043] Among them, the queue ratio is the ratio of the vehicle queue length to the road length, which is used to measure the road congestion degree. The vehicle accumulation rate is the ratio of the vehicle arrival rate to the departure rate, which is used to measure the traffic flow backlog degree. The preset state threshold includes a first warning threshold and a second warning threshold, which are used to divide the traffic state levels. The normal state indicates that the traffic flow runs smoothly. The warning state indicates that there is a congestion trend in the traffic flow. The dangerous state indicates that traffic congestion has formed. The state division period is usually set to 60 seconds.

[0044] This step is executed after traffic data collection and is used to evaluate the traffic operation status. Specifically, the system first obtains the total road length L0 and the current queue length L1 of each approach lane, and calculates the queue ratio r1 = L1 / L0. Then it obtains the vehicle arrival rate f1 and the departure rate f2, and calculates the vehicle accumulation rate r2 = f1 / f2. The calculated r1 is compared with the first warning threshold of 0.6 and the second warning threshold of 0.8, and r2 is compared with the first accumulation rate threshold of 1.2 and the second accumulation rate threshold of 1.5. When r1 < 0.6 and r2 < 1.2, it is determined to be in a normal state; when 0.6 ≤ r1 < 0.8 or 1.2 ≤ r2 < 1.5, it is determined to be in a warning state; when r1 ≥ 0.8 or r2 ≥ 1.5, it is determined to be in a dangerous state.

[0045] In some embodiments, the calculation and classification of traffic states can be achieved in various ways: Optionally, a fuzzy evaluation scheme can be used: establish fuzzy sets of queue ratio and accumulation rate, design membership functions, calculate the membership values of each state, and determine the traffic state according to the principle of maximum membership; Optionally, a comprehensive index scheme can be used: introduce supplementary indicators such as vehicle speed and occupancy rate, obtain a comprehensive score through weighted summation, set score intervals corresponding to different traffic states, and achieve state classification. It can be understood that other methods such as neural network evaluation and expert system can also be used to determine the traffic state, which is not limited here.

[0046] S103. When the traffic state is in the warning state, generate a traffic flow time series curve according to the real-time traffic data.

[0047] Among them, the warning state refers to the traffic operation state where the queue ratio exceeds the first warning threshold but does not reach the second warning threshold. Real-time traffic data refers to the data set reflecting the current traffic operation state obtained through detection devices. The traffic flow time series curve refers to a two-dimensional curve graph with time as the abscissa and traffic flow as the ordinate, which is used to represent the change trend of traffic flow over time. Time series refers to a data sequence arranged in chronological order. Generation refers to obtaining a continuous change curve through data processing and curve fitting. The data sampling period is usually set to 20 seconds, and the data storage duration is one complete signal cycle.

[0048] This step is executed when the system determines that the traffic state is in the warning state and is used to analyze the traffic flow change law. Specifically, the system first confirms that the approach lane is in the warning state and reads the traffic flow data within the most recent complete signal cycle. For each sampling moment, calculate the number of vehicles passing through the detector per unit time to obtain discrete flow data points. Use the cubic spline interpolation method to fit these discrete data points to generate a smooth traffic flow time series curve. The abscissa range of the curve is from 0 to the signal cycle length, and the ordinate range is from 0 to the upper limit of road capacity.

[0049] In some embodiments, the generation of the traffic flow time series curve can be achieved in various ways: Optionally, the direct drawing scheme is used: collect the traffic flow data at each sampling moment, connect the adjacent sampling points with straight line segments to form a piecewise linear time series curve, and calculate the mean and standard deviation of the curve to evaluate the degree of data dispersion; Optionally, the curve fitting scheme is used: perform data preprocessing on the sampling data to remove outliers, calculate the polynomial fitting coefficients using the least squares method, generate a smooth continuous curve, and calculate the goodness-of-fit evaluation index. It can be understood that other methods such as wavelet analysis and Fourier transform can also be used to generate the traffic flow time series curve, which is not limited here.

[0050] S104. Extract traffic flow characteristic parameters from the traffic flow time series curve. The traffic flow characteristic parameters include the maximum traffic flow, the minimum traffic flow, the time when the maximum value appears, and the time when the minimum value appears.

[0051] Among them, the traffic flow time series curve represents a two-dimensional graph of the traffic flow changing with time. The traffic flow characteristic parameters refer to a set of key indicators that describe the operating characteristics of the traffic flow. The maximum traffic flow refers to the flow value corresponding to the highest point on the curve. The minimum traffic flow refers to the flow value corresponding to the lowest point on the curve. The time when the maximum value appears refers to the time point corresponding to when the traffic flow reaches the maximum value. The time when the minimum value appears refers to the time point corresponding to when the traffic flow reaches the minimum value. Extraction refers to the process of obtaining characteristic parameters through numerical calculation. The parameter storage period is usually set to a complete signal cycle.

[0052] This step is executed after the generation of the traffic flow time series curve to obtain the key characteristics of the traffic flow. Specifically, the system first performs numerical analysis on the traffic flow time series curve, and determines the extreme points of the curve through derivative calculation. Calculate the first derivative of the curve over the entire period. When the derivative changes from positive to negative, it is a maximum point, and when the derivative changes from negative to positive, it is a minimum point. Record the flow values and corresponding times of all extreme points, compare the flow values of all maximum points, and select the largest one as the maximum traffic flow, and the corresponding time as the time when the maximum value appears. Compare the flow values of all minimum points, and select the smallest one as the minimum traffic flow, and the corresponding time as the time when the minimum value appears.

[0053] In some embodiments, the extraction of traffic flow characteristic parameters can be achieved in various ways: Optionally, a numerical differentiation scheme can be used: calculate the left and right derivatives of each point on the curve, determine the extreme points by the change of the derivative sign, confirm the type of extreme value in combination with the second derivative, record the positions and values of the extreme points, and screen the maximum and minimum values therefrom; Optionally, a sliding window scheme can be used: set an appropriate window length, slide on the curve to search for local maximum and minimum values, compare all local extreme values to obtain the global maximum and minimum values, and record the corresponding time information. It can be understood that other methods such as statistical analysis and peak detection can also be used to achieve the extraction of traffic flow characteristic parameters, which are not limited herein.

[0054] S105. Arrange the traffic flow characteristic parameters according to the number of the import lane to construct a traffic flow characteristic matrix.

[0055] Among them, the traffic flow characteristic parameters refer to a set of key indicators that describe the operating characteristics of traffic flow, including the maximum and minimum traffic volumes and their corresponding times. The import lane number refers to the unique identifier of each import lane at the intersection, usually numbered clockwise according to the geographical location. The traffic flow characteristic matrix refers to a two-dimensional data table arranged in the order of the import lane numbers, used to store the traffic flow characteristic parameters of each import lane. Matrix construction refers to the process of organizing discrete characteristic parameters into structured data according to specific rules. Arrangement refers to the data sorting in ascending or descending order according to the import lane numbers. The number of rows of the matrix is equal to the number of import lanes, and the number of columns is equal to the number of characteristic parameters.

[0056] This step is executed after the extraction of traffic flow characteristic parameters and is used to construct the basis for data analysis. Specifically, the system first determines the dimensions of the matrix. The number of rows is equal to the number of import lanes n, and the number of columns is equal to the number of characteristic parameters 4. Create an empty matrix of n×4. Each row of the matrix corresponds to an import lane, and each column corresponds to the maximum traffic volume, the maximum time, the minimum traffic volume, and the minimum time respectively. According to the ascending order of the import lane numbers, fill the traffic flow characteristic parameters of each import lane into the corresponding positions of the matrix. Perform normalization processing on the matrix, divide the traffic volume value by the upper limit of the road capacity, and divide the time value by the signal cycle length.

[0057] In some embodiments, the construction of the traffic flow characteristic matrix can be achieved in various ways: Optionally, a sequential filling scheme can be used: create a two-dimensional array to store the matrix, read the characteristic parameters in the order of the import lane numbers, fill the parameters into the corresponding positions of the array in sequence, and perform data verification on the filled matrix to ensure data integrity and correctness; Optionally, a parallel processing scheme can be used: process the characteristic parameters of multiple import lanes simultaneously, create a temporary data cache, determine the position of the data in the matrix according to the import lane numbers, write the data concurrently, and finally perform data synchronization and verification. It can be understood that other methods such as sparse matrix storage and database table storage can also be used to achieve the construction of the traffic flow characteristic matrix, which are not limited herein.

[0058] S106. Analyze the traffic flow complementary relationship between import lanes based on the traffic flow characteristic matrix, and calculate the phase compensation parameter according to the traffic flow complementary relationship.

[0059] Among them, the traffic flow characteristic matrix refers to a two-dimensional data table storing the traffic flow characteristic parameters of each import lane. The traffic flow complementary relationship refers to the property of the traffic flow change between different import lanes being offsetting. The phase compensation parameter refers to the time correction amount used to adjust the signal phase difference. Analysis refers to the process of studying the correlation between data through mathematical methods. Calculation refers to the process of obtaining numerical results through specific algorithms. The complementary relationship is usually quantified by the correlation coefficient, and the value range of the correlation coefficient is [-1, 1].

[0060] This step is executed after the construction of the traffic flow characteristic matrix to determine the import lane timing optimization strategy. Specifically, the system first calculates the traffic flow correlation coefficient between any two import lanes. For import lanes i and j, extract their traffic flow time series data and calculate the Pearson correlation coefficient rij. When rij < -0.6, it is considered that the two import lanes have a significant complementary relationship. After determining the import lane combination with the complementary relationship, calculate the difference between the maximum traffic flow moment t1 of the first import lane and the minimum traffic flow moment t2 of the second import lane, and use the time difference as the basic value of the phase compensation parameter. Weight and correct the basic value according to the traffic flow maximum value ratio of the two import lanes to obtain the final phase compensation parameter.

[0061] In some embodiments, the complementary relationship analysis and phase compensation parameter calculation can be implemented in various ways: Optionally, use the correlation analysis scheme: perform data preprocessing on the traffic flow characteristic matrix, calculate the correlation coefficient matrix between import lanes, set the correlation threshold to screen the import lane pairs with significant complementary relationships, and calculate the basic compensation value according to the time series characteristics; Optionally, use the clustering analysis scheme: perform spatio-temporal clustering on the traffic flow data, identify the import lane groups with complementary change patterns, calculate the inter-class distance as a measure of the complementary degree, and determine the compensation parameter in combination with the flow ratio. It can be understood that other methods such as time series analysis and neural networks can also be used to implement the complementary relationship analysis and phase compensation parameter calculation, which are not limited here.

[0062] S107. Adjust the signal timing sequence according to the phase compensation parameter.

[0063] Among them, the phase compensation parameter refers to the time correction amount used to adjust the signal phase difference, expressed in seconds. The signal timing sequence refers to the time schedule of each phase within a complete cycle, including parameters such as phase difference and green light duration. Adjustment refers to the process of modifying the original timing plan according to the phase compensation parameter. The signal cycle refers to the time required for a complete set of signal displays to be completed. The phase difference refers to the time difference between the starting moments of the green lights at adjacent approaches. The phase green light duration refers to the duration during which the signal light shows green. The adjustment accuracy of the timing parameters is usually set to 1 second. The cycle length is usually between 60 seconds and 180 seconds.

[0064] This step is executed after calculating the phase compensation parameter and is used to optimize the signal timing plan. Specifically, the system first reads the current signal timing sequence, including the cycle length T, the phase differences Di and the green light durations Gi of each approach. For the approach pair (i, j) with a complementary relationship, subtract the phase compensation parameter P from the phase difference of approach i to obtain the new phase difference Di_new = Di - P. If Di_new is less than 0, add the cycle length T; if Di_new is greater than T, subtract the cycle length T to ensure that the new phase difference is within the range of [0, T]. Keep the green light durations of each approach unchanged and update the phase difference parameter in the timing sequence. Send the modified timing sequence to the signal controller for execution.

[0065] In some embodiments, the adjustment of the signal timing sequence can be achieved in various ways: Optionally, use the direct adjustment scheme: Read the original timing sequence, obtain the phase compensation parameter, directly modify the phase differences of the complementary approaches, perform cycle normalization processing, verify whether the adjusted timing plan meets the minimum green light duration constraint, and generate a new timing sequence; Optionally, use the optimized adjustment scheme: Establish a timing optimization model, use the phase compensation parameter as a constraint condition, consider the goal of minimizing delay, and use a genetic algorithm to solve the optimal timing plan, and conduct a feasibility test on the solution result. It can be understood that other methods such as fuzzy control and dynamic programming can also be used to achieve the adjustment of the signal timing sequence, which is not limited here.

[0066] The following further describes the more specific process of the method provided in this embodiment. Please refer to Figure 2 , which is another process schematic diagram of the queuing overflow control method in the second-level adaptive control system in the embodiment of the present application.

[0067] S201. Arrange the vehicle flow feature matrix in chronological order to form a time series matrix.

[0068] The traffic flow feature matrix is a 4×4 matrix, where each row contains the maximum flow value, minimum flow value, maximum value time, and minimum value time of an approach. The time series matrix is a matrix obtained by rearranging the traffic flow features according to time, reflecting the variation law of the traffic flow of each approach over time. The time order refers to the arrangement in the chronological order of time occurrence. The system reorganizes the traffic flow feature matrix into a time series matrix. First, the maximum value time and minimum value time of each approach are extracted, and all the times are sorted from smallest to largest. According to the sorted time series, the system constructs an n×4 matrix (n is the total number of times), where each row of the matrix corresponds to a time point, and the four columns of data represent the traffic flows of the four approaches at that time. For the maximum value time, the corresponding maximum flow value is filled in; for the minimum value time, the corresponding minimum flow value is filled in; for other time points, the corresponding flow value is calculated by linear interpolation. The obtained time series matrix completely records the variation process of the traffic flow of each approach over time.

[0069] S202. Calculate the traffic flow correlation coefficient between any two approaches based on this time series matrix.

[0070] The traffic flow correlation coefficient is a statistical index measuring the degree of association of traffic flow changes between two approaches, with a value range of [-1, 1]. A positive correlation coefficient indicates a positive correlation, a negative one indicates a negative correlation, and the larger the absolute value, the stronger the correlation. The time series matrix contains the traffic flow data of each approach at different times. The system uses the Pearson correlation coefficient method to calculate the correlation between any two approaches. For approaches i and j, two corresponding columns of data are extracted from the time series matrix, denoted as Xi and Xj. Calculate the covariance cov(Xi, Xj) and standard deviations σi, σj of these two columns of data, and then calculate the correlation coefficient according to the formula ρij = cov(Xi, Xj) / (σi×σj). Specifically, the covariance calculation formula is cov(Xi, Xj) = Σ[(Xi - μi)(Xj - μj)] / n, where μi and μj are the averages of Xi and Xj respectively, and n is the sample size. The standard deviation calculation formula is σ = √[Σ(X - μ)² / n]. The system calculates the correlation coefficient for all possible approach combinations to form a 4×4 correlation coefficient matrix.

[0071] S203. Determine the approach combinations with a traffic flow correlation coefficient less than the preset threshold as the approach combinations with complementary relationships.

[0072] The traffic flow correlation coefficient reflects the synchronization of traffic flow changes between two approach roads. The preset threshold is the standard value for judging the complementary relationship, usually set as a negative value. The complementary relationship means that the traffic flows of two approach roads show a characteristic of one increasing while the other decreasing, that is, when the traffic flow of one approach road increases, the traffic flow of the other approach road decreases. The system sets the preset threshold to -0.6, indicating that when the correlation coefficient between two approach roads is less than -0.6, they are considered to have a significant complementary relationship. The system traverses each non-diagonal element in the correlation coefficient matrix and records the pairs of approach roads that meet the conditions. For each pair of approach roads with a complementary relationship, the system records their number pairs. For example, (1, 3) means that approach road No. 1 and approach road No. 3 have a complementary relationship. If there are multiple combinations that meet the conditions, the system stores them in the complementary relationship list. These pairs of approach roads with a complementary relationship will be used for subsequent calculation of phase compensation parameters.

[0073] S204. Calculate the time interval between the moment of the maximum traffic flow of the first approach road and the moment of the minimum traffic flow of the second approach road in the pair of approach roads with a complementary relationship.

[0074] The time interval refers to the duration difference between two time points. The moment of the maximum traffic flow refers to the time point when the traffic flow of a certain approach road reaches the peak within the traffic signal cycle. The moment of the minimum traffic flow refers to the time point when the traffic flow of a certain approach road reaches the trough within the traffic signal cycle. The pair of approach roads with a complementary relationship refers to two approach roads whose correlation coefficient is less than the preset threshold, where the one with the smaller number is the first approach road and the one with the larger number is the second approach road. The calculation reference point for the moment is the start moment of a complete signal cycle, that is, the 0 moment. The system extracts the moment of the maximum traffic flow of the first approach road from the time series matrix and records it as T1, and the moment of the minimum traffic flow of the second approach road and records it as T2, and performs the subtraction operation T1 - T2 to obtain the time interval ΔT. This time interval reflects the time relationship between the traffic flow peaks and troughs of the approach roads with a complementary relationship and provides a basis for subsequent phase difference optimization.

[0075] S205. Use this time interval as the initial value of this phase compensation parameter; perform weighted correction on the initial value of the phase compensation parameter according to the ratio of the maximum traffic flow of the first approach road to the minimum traffic flow of the second approach road to obtain the phase compensation parameter.

[0076] The phase compensation parameter is a time parameter used to adjust the phase difference in traffic signal timing, and its initial value is derived from the time interval between the peak and valley of traffic flow. The ratio of the maximum traffic flow to the minimum traffic flow reflects the relative change degree of traffic flow intensity. Weighted correction is the process of adjusting the initial compensation parameter according to the flow ratio. In a complete signal cycle, the phase corresponds to the conversion timing of the signal lights, and the phase compensation parameter directly affects the adjustment amount of the green light start time. The system first sets the time interval ΔT as the initial value P0 of the phase compensation parameter, and then calculates the ratio r = M1 / M2 of the maximum traffic flow M1 of the first approach to the minimum traffic flow M2 of the second approach. When r > 2, the system multiplies the initial value by 1.2 to obtain the final compensation parameter; when 1 ≤ r ≤ 2, the initial value is directly adopted; when r < 1, the initial value is divided by 1.2. This weighted correction method comprehensively considers the influence of time difference and flow ratio.

[0077] S206. Subtract the phase compensation parameter from the current phase difference to obtain the target phase difference.

[0078] The current phase difference refers to the time difference between the green light start times of two complementary approach lanes in the existing signal timing plan. The target phase difference is the time difference between the green light start times of two approach lanes after optimization. The phase compensation parameter is the time amount that needs to be adjusted. In traffic signal timing, the correspondence between the phase and the traffic light time is: the start point of the phase corresponds to the green light start time, the end point of the phase corresponds to the green light end time, and the phase length is equal to the green light duration. The system reads the phase difference D from the current timing plan, subtracts the calculated phase compensation parameter P, and obtains the target phase difference Dnew = D - P. If the calculation result exceeds the range of the signal cycle length T, cycle normalization is required: when Dnew < 0, add the cycle length T; when Dnew > T, subtract the cycle length T to ensure that the target phase difference is within the effective range.

[0079] S207. Adjust the difference between the phase start times of the first approach lane and the second approach lane in the complementary approach lane combination to the target phase difference.

[0080] The phase start time refers to the moment when the signal lights of each approach lane change from red to green. The phase difference refers to the time difference between the phase start times of two approach lanes. The target phase difference is the optimized phase difference calculated through the traffic flow complementary relationship. The relationship between the phase time and the traffic light time is: the phase start time corresponds to the green light start time, the phase end time corresponds to the green light end time, and the phase duration is the green light duration. When the system adjusts the phase difference, it keeps the signal cycle length unchanged and realizes the adjustment of the phase difference by changing the green light start time.

[0081] The system adjusts the phases of the import lane combinations with complementary relationships. First, it reads the current phase start times t1 and t2 of the first import lane and the second import lane, and calculates the current phase difference dc = t1 - t2. Then, it adjusts the phase difference to the target phase difference dt. The adjustment method is as follows: keep the phase start time t1 of the first import lane unchanged, and adjust the phase start time of the second import lane to t1 - dt. If the adjusted time exceeds the cycle length T, perform cycle normalization: when the time value is greater than T, subtract T; when the time value is less than 0, add T. This ensures that the adjusted phase start time is within the range of [0, T].

[0082] S208. Set the first green light phase duration according to the maximum traffic flow value of this first import lane, and set the second green light phase duration according to the ratio of the maximum traffic flow value to the minimum traffic flow value of this second import lane.

[0083] The green light phase duration refers to the duration when the signal light shows green. The maximum traffic flow value reflects the maximum traffic demand of the import lane. The ratio of the maximum traffic flow value to the minimum traffic flow value represents the degree of traffic demand fluctuation. The first green light phase duration and the second green light phase duration respectively correspond to the green light durations of the first import lane and the second import lane. The setting of the phase duration directly affects the traffic capacity of each import lane.

[0084] The system sets the green light phase duration according to the traffic flow data. For the first import lane, substitute its maximum traffic flow value Q1max into the formula t1 = 20 + Q1max / 50 (unit: seconds) to calculate the first green light phase duration. For the second import lane, first calculate the ratio r = Q2max / Q2min of its maximum traffic flow value to the minimum traffic flow value, and then substitute it into the formula t2 = 15 + Q2max / 50 + 5r (unit: seconds) to calculate the second green light phase duration. This calculation method takes into account both the influence of the absolute value of the traffic flow and the factor of traffic flow fluctuation.

[0085] S209. Allocate the remaining duration to the green light phases of the third import lane and the fourth import lane according to the ratio of the maximum traffic flow value.

[0086] The remaining duration refers to the remaining time in the signal cycle after deducting the green light phase durations of the first import lane and the second import lane. The third import lane and the fourth import lane refer to the other two import lanes not included in the complementary relationship combination. Phase allocation means allocating the available time to each import lane as its green light display time. The ratio of the maximum traffic flow value reflects the relative traffic demands of each import lane.

[0087] The system calculates the remaining duration \(t_r = T - t_1 - t_2 - t_y\), where \(T\) is the cycle length, \(t_1\) and \(t_2\) are the set green - phase durations, and \(t_y\) is the yellow - phase duration. Then it obtains the maximum traffic flows \(Q_{3max}\) and \(Q_{4max}\) of the third and fourth approach lanes, and calculates the proportion coefficients \(p_3=Q_{3max} / (Q_{3max}+Q_{4max})\) and \(p_4 = Q_{4max} / (Q_{3max}+Q_{4max})\). The remaining duration is distributed proportionally: the green - phase duration of the third approach lane \(t_3=t_r\times p_3\), and the green - phase duration of the fourth approach lane \(t_4=t_r\times p_4\). This distribution method ensures that each approach lane obtains a green - light time suitable for its traffic demand.

[0088] S210. Determine the red - phase duration of each such approach lane according to the difference in the starting moments of the phases. The sum of the green - phase duration and the red - phase duration of each such approach lane is equal to the total cycle duration.

[0089] The difference in the starting moments of the phases refers to the time difference between the starting moments of the green lights of adjacent approach lanes. The red - phase duration refers to the duration during which the signal shows red. The green - phase duration refers to the duration during which the signal shows green. The total cycle duration refers to the time required for a complete signal cycle, including all red - light durations, green - light durations, and yellow - light durations. In the phase time sequence, the red - phase duration and the green - phase duration are complementary sets, and their sum plus the yellow - light duration is equal to the total cycle duration. The system calculates the red - phase duration of each approach lane according to the difference in the starting moments of the phases determined in the foregoing steps. Let the green - phase duration of approach lane \(i\) be \(G_i\), and the difference in the starting moments of the phases of the subsequent approach lane be \(D_i\). Then the red - phase duration \(R_i\) of approach lane \(i\) is \(R_i = T - G_i - Y\), where \(T\) is the total cycle duration and \(Y\) is the yellow - light duration. This calculation method ensures the continuity of the signal timing of adjacent approach lanes and the integrity of the cycle.

[0090] S211. When the traffic state is in the normal state, obtain the vehicle - gap acceptance rate of the approach lane.

[0091] The vehicle gap refers to the time interval between two consecutive passing vehicles. The gap acceptance rate refers to the proportion of all vehicle gaps that are greater than the critical gap. The normal state refers to the traffic state in which both the queue proportion and the vehicle accumulation rate do not exceed the first - level threshold. The critical gap is usually set to 2.5 seconds, which represents the minimum headway required for safe and efficient passage. The system obtains the time - interval data of consecutive passing vehicles of each approach lane through vehicle detectors. For \(n\) consecutive passing vehicles, record \(n - 1\) vehicle - gap values, count the number \(m\) of those greater than the critical gap, and calculate the gap acceptance rate \(r=m / (n - 1)\). In this way, the system obtains a quantitative index reflecting the density of the traffic flow.

[0092] S212. When the vehicle gap acceptance rate is greater than the preset acceptance rate threshold, shorten the signal timing sequence by a preset time length.

[0093] The preset acceptance rate threshold is the standard value for judging whether the traffic flow is too sparse, usually set to 0.3. The preset time length is the basic unit for signal timing adjustment, usually set to 5 seconds. The signal timing sequence refers to the time arrangement of each phase within a complete signal cycle. Shortening the timing sequence means reducing the duration of each phase proportionally, thereby reducing the total cycle duration. The system first compares the vehicle gap acceptance rate with the preset threshold of 0.3. When the acceptance rate is greater than 0.3, it indicates that the current traffic flow is relatively sparse. At this time, subtract the preset time length of 5 seconds from the current cycle duration T to obtain the new cycle duration Tnew = T - 5. The duration of each phase is reduced in proportion, that is, the new phase duration is equal to the original phase duration multiplied by (Tnew / T). This adjustment method maintains the duration ratio relationship between each phase and improves the efficiency of signal control at the same time.

[0094] S213. When the vehicle gap acceptance rate is not greater than the preset acceptance rate threshold, extend the signal timing sequence by a preset time length.

[0095] The vehicle gap acceptance rate refers to the proportion of the time when the vehicle gap is greater than the critical gap. The preset acceptance rate threshold is the standard value for judging the density of the traffic flow, set to 0.3. The preset time length is the basic unit for signal timing adjustment, set to 5 seconds. The signal timing sequence includes the start and end times and the duration of each phase within a complete cycle. Extending the timing sequence means increasing the duration of each phase proportionally and increasing the total cycle duration. When the vehicle gap acceptance rate is not greater than 0.3, the system increases the existing cycle duration T by 5 seconds to obtain the new cycle duration Tnew = T + 5. The duration of each phase is increased in proportion, and the new phase duration is equal to the original phase duration multiplied by (Tnew / T). This adjustment keeps the duration ratio of each phase unchanged and adapts to the relatively dense traffic flow state by increasing the total cycle duration.

[0096] S214. Monitor the change trends of the queue proportion and the vehicle accumulation rate in real time.

[0097] The queuing ratio refers to the ratio of the queuing length at the entrance lane to the road length. The vehicle accumulation rate refers to the ratio of the number of vehicles entering the intersection per unit time to the number of vehicles leaving the intersection. The changing trend refers to the direction and rate of change of these two parameters over time. The monitoring period refers to the time interval for the system to collect and update data, which is set to 20 seconds. The system collects data through vehicle detectors every 20 seconds and calculates the latest queuing ratio and vehicle accumulation rate. The calculation method of the queuing ratio is to divide the current queuing length by the total road length. The calculation method of the vehicle accumulation rate is to divide the number of vehicles passing through the upstream detector within 20 seconds by the number of vehicles passing through the downstream detector. The system stores these data in chronological order for subsequent trend analysis.

[0098] S215. When it is detected that the queuing ratio continuously rises in multiple consecutive signal cycles and exceeds the first warning threshold, switch the traffic state to the warning state.

[0099] The first warning threshold is the standard value of the queuing ratio for triggering the warning state, which is set to 0.6. The warning state refers to the state where traffic congestion begins to worsen but has not reached a severe level. Multiple consecutive signal cycles usually refer to 3 or more signal cycles. The state switch refers to the change of the system operation mode from the normal state to the warning state. The system checks the value of the queuing ratio at the end of each signal cycle. When it is detected that the queuing ratio shows an upward trend in 3 consecutive cycles and the latest value exceeds 0.6, the system changes the operation state flag from 1 (normal state) to 2 (warning state). In the warning state, the system will initiate corresponding traffic guidance strategies, including measures such as adjusting the phase difference and green light duration.

[0100] S216. When the traffic state is in the dangerous state, calculate the vehicle saturation degree of each entrance lane. The vehicle saturation degree is equal to the product of the vehicle arrival rate and the vehicle gap acceptance rate.

[0101] The vehicle saturation degree is an indicator to measure the matching degree between traffic demand and traffic capacity. The vehicle arrival rate is the ratio of the number of vehicles arriving at the intersection per unit time to the maximum number of vehicles that the road can accommodate. The vehicle gap acceptance rate is the time ratio greater than the critical gap. The dangerous state refers to the traffic state where the queuing ratio or the vehicle accumulation rate exceeds the second-level threshold. The method for the system to calculate the vehicle saturation degree is as follows: First, obtain the number of vehicles N1 passing through the upstream detector within 20 seconds for each entrance lane, divide N1 by the maximum number of vehicles N0 that the entrance lane can accommodate to get the vehicle arrival rate r1 = N1 / N0; then count the total length t1 of the time periods when the vehicle intervals are greater than 2.5 seconds, divide it by the observation duration of 20 seconds to get the vehicle gap acceptance rate r2 = t1 / 20; finally, multiply r1 by r2 to get the vehicle saturation degree S = r1 × r2. This calculation method takes into account both traffic demand and traffic capacity.

[0102] S217. Prioritize the import lane according to the vehicle saturation of this lane.

[0103] The priority is the service order level of the import lane determined according to the vehicle saturation. The prioritization is to arrange the import lanes in descending order of vehicle saturation. The sorting result directly determines the order in which each import lane obtains the additional green light time. The system first creates a two-dimensional array containing the import lane numbers and the corresponding vehicle saturations. Each row of the array contains two elements: the import lane number and the vehicle saturation. Use the quicksort algorithm to sort this array in descending order of vehicle saturation to obtain the priority order. The first row of the sorted array corresponds to the import lane with the highest priority, and the last row corresponds to the import lane with the lowest priority. This sorting method ensures that the import lane with the greatest traffic flow pressure is given priority for dredging.

[0104] S218. Extend the green light period of the import lane with the highest priority to a preset time.

[0105] The import lane with the highest priority refers to the import lane with the greatest vehicle saturation. The green light period refers to the time period when the signal light shows green. The preset time is the fixed duration for extending the green light, which is set to 15 seconds. The extended period is the time added on the basis of the original green light duration. The system first reads the number of the first import lane in the priority sorting result and obtains the current green light duration t1 of this import lane. Add 15 seconds to t1 to get the new green light duration t2 = t1 + 15. To keep the total cycle length unchanged, the green light durations of other import lanes are reduced proportionally, and the sum of the reduction amounts is equal to 15 seconds. The specific calculation method is: the new green light duration of other import lanes is equal to the original green light duration multiplied by [(T - t2) / (T - t1)], where T is the total cycle length. This adjustment method gives priority to ensuring the traffic demand of the import lane with the greatest traffic flow pressure.

[0106] S219. When the queuing ratio drops below the second warning threshold and the vehicle accumulation rate shows a downward trend, and at the same time the vehicle gap acceptance rate is within the preset normal range, switch the traffic state to the warning state.

[0107] The queuing ratio is the ratio of the queuing length of the import lane to the road length, which is calculated in real time by the vehicle detector. The second warning threshold is the standard value of the queuing ratio for judging the traffic state to change from dangerous to warning, which is set to 0.8. The vehicle accumulation rate is the ratio of the number of vehicles entering the intersection to the number of vehicles leaving the intersection, and the downward trend means that this value continuously decreases for 3 consecutive cycles. The vehicle gap acceptance rate is the proportion of the time greater than the critical gap, and the normal range refers to between 0.2 and 0.4. The warning state is the intermediate state of traffic operation, between the normal state and the dangerous state. The state switch means that the system operation mode changes from the dangerous state to the warning state.

[0108] The system performs a status check every 20 seconds. First, it calculates the current queue length L1 through the upstream detector, divides L1 by the total road length L0 to obtain the queue ratio r1 = L1 / L0, and checks if r1 is less than 0.8. Then, it calculates the vehicle accumulation rates for the last 3 cycles, denoted as c1, c2, and c3 respectively, and checks if c3 < c2 < c1. Next, it calculates the total length t1 of time periods when the vehicle interval is greater than 2.5 seconds, divides t1 by the observation duration of 20 seconds to obtain the vehicle gap acceptance rate r2 = t1 / 20, and checks if r2 is between 0.2 and 0.4. When all three conditions are met simultaneously, the system changes the operation status flag from 3 (hazardous state) to 2 (warning state). Along with the status switch, the system resumes the traffic control strategies in the warning state, including measures such as phase difference optimization and signal timing adjustment. This multi-condition judgment mechanism ensures the stability and reliability of the status switch.

[0109] The following describes the second-level adaptive control system in the embodiments of the present invention application from the perspective of hardware processing. Please refer to Figure 3 , which is a schematic structural diagram of an entity device of the second-level adaptive control system in the embodiments of the present application.

[0110] It should be noted that Figure 3 the structure of the second-level adaptive control system shown is only an example and should not impose any limitations on the functions and usage scope of the embodiments of the present invention.

[0111] As Figure 3 shown, the second-level adaptive control system includes a Central Processing Unit (CPU) 301, which can perform various appropriate actions and processes according to the program stored in the Read-Only Memory (ROM) 302 or the program loaded from the storage section 308 into the Random Access Memory (RAM) 303, such as executing the methods described in the above embodiments. In the RAM 303, various programs and data required for system operation are also stored. The CPU 301, ROM 302, and RAM 303 are connected to each other through a bus 304. The Input / Output (I / O) interface 305 is also connected to the bus 304.

[0112] The following components are connected to the I / O interface 305: an input section 306 including an audio input device, a button switch, etc.; an output section 307 including a liquid crystal display (LCD), an audio output device, an indicator light, etc.; a storage section 308 including a hard disk, etc.; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card, a modem, etc. The communication section 309 performs communication processing via a network such as the Internet. The drive 310 is also connected to the I / O interface 305 as needed. A removable medium 311, such as a magnetic disk, an optical disk, a magneto-optical disk, a semiconductor memory, etc., is installed on the drive 310 as needed so that a computer program read from it can be installed into the storage section 308 as needed.

[0113] Specifically, according to an embodiment of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, an embodiment of the present invention includes a computer program product, which includes a computer program carried on a computer-readable medium, and the computer program contains a computer program for executing the method shown in the flowchart. In such an embodiment, the computer program can be downloaded and installed from the network through the communication section 309, and / or installed from the removable medium 311. When the computer program is executed by the central processing unit (CPU) 301, various functions defined in the present invention are executed.

[0114] It should be noted that specific examples of computer-readable storage media may include, but are not limited to: an electrical connection having one or more wires, a portable computer disk, a hard disk, a random access memory (RAM), a read-only memory (ROM), an erasable programmable read-only memory (EPROM), a flash memory, an optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination of the above. In the present invention, a computer-readable storage medium can be any tangible medium that contains or stores a program that can be used by or in conjunction with an instruction execution system, apparatus, or device.

[0115] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of systems, methods, and computer program products according to various embodiments of the present invention. Among them, each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the above-mentioned module, program segment, or part of code contains one or more executable instructions for implementing the specified logical function. It should also be noted that in some alternative implementations, the functions marked in the block may occur in a different order than that marked in the accompanying drawings.

[0116] Specifically, the second-level adaptive control system of this embodiment includes a processor and a memory. A computer program is stored on the memory. When the computer program is executed by the processor, it implements the queuing overflow control method in the second-level adaptive control system provided in the above embodiment.

[0117] On the other hand, the present invention also provides a computer-readable storage medium. This storage medium may be included in the second-level adaptive control system described in the above embodiment; or it may exist separately and not be assembled into the second-level adaptive control system. The above storage medium carries one or more computer programs. When the above one or more computer programs are executed by a processor of the second-level adaptive control system, the second-level adaptive control system implements the queuing overflow control method in the second-level adaptive control system provided in the above embodiment.

[0118] As mentioned above, the above embodiments are only used to illustrate the technical solutions of the present application, rather than to limit them; although the present application has been described in detail with reference to the foregoing embodiments, those of ordinary skill in the art should understand that: they can still modify the technical solutions recorded in the foregoing embodiments, or perform equivalent replacements on some of the technical features; and these modifications or replacements do not make the essence of the corresponding technical solutions deviate from the scope of the technical solutions of the various embodiments of the present application.

[0119] As used in the above embodiments, depending on the context, the term "when..." may be interpreted to mean "if...", or "after...", or "in response to determining...", or "in response to detecting...". Similarly, depending on the context, the phrase "when determining..." or "if detecting (the stated condition or event)" may be interpreted to mean "if determining...", or "in response to determining...", or "when detecting (the stated condition or event)", or "in response to detecting (the stated condition or event)".

[0120] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by relevant hardware instructed by a computer program. This program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the above method embodiments. The foregoing storage medium includes: various media that can store program codes such as ROM, random access memory (RAM), magnetic disks, or optical discs.

Claims

1. A queue overflow control method in a second-level adaptive control system, characterized in that: Applied to a second-level adaptive control system, the method comprises: Collecting real-time traffic data of each entrance road, wherein the real-time traffic data includes vehicle queue length, vehicle arrival rate and vehicle departure rate; Calculating the queue ratio and vehicle accumulation rate of each of the entrance lanes according to the real-time traffic data, and dividing the traffic state of each of the entrance lanes into a normal state, a warning state, and a dangerous state according to a preset state threshold; When the traffic state is in the warning state, generating a traffic flow time series curve according to the real-time traffic data; Extracting traffic flow characteristic parameters from the traffic flow time series curve, wherein the traffic flow characteristic parameters include a maximum traffic flow value, a minimum traffic flow value, a maximum value occurrence time, and a minimum value occurrence time; Arranging the traffic flow characteristic parameters according to the numbers of the entrance lanes to construct a traffic flow characteristic matrix; Arranging the traffic flow feature matrix in chronological order to form a time series matrix; Calculate the traffic flow correlation coefficient between any two entrance lanes according to the time series matrix; Determining the inlet lane combination whose vehicle flow correlation coefficient is less than a preset threshold as an inlet lane combination having a complementary relationship; Calculating the time interval between the maximum value of the traffic flow of the first entrance road and the minimum value of the traffic flow of the second entrance road in the entrance road combination having a complementary relationship; The time interval is used as an initial value of a phase compensation parameter; the initial value of the phase compensation parameter is weightedly modified according to a ratio of a maximum vehicle flow rate of the first entrance road to a minimum vehicle flow rate of the second entrance road to obtain a phase compensation parameter; The signal timing sequence is adjusted according to the phase compensation parameter.

2. The method according to claim 1, characterized in that The step of adjusting the signal timing sequence according to the phase compensation parameter specifically includes: Subtract the phase compensation parameter from the current phase difference to obtain a target phase difference; Adjusting the difference between the phase start time of the first inlet and the phase start time of the second inlet in the inlet combination having a complementary relationship to the target phase difference; The first green light phase duration is set according to the maximum value of the traffic flow of the first entrance road, and the second green light phase duration is set according to the ratio of the maximum value to the minimum value of the traffic flow of the second entrance road; the remaining duration is allocated to the green light phases of the third entrance road and the fourth entrance road according to the proportion of the maximum value of the traffic flow; The red light phase duration of each entrance lane is determined according to the difference between the phase start times, and the sum of the green light phase duration and the red light phase duration of each entrance lane is equal to the total cycle duration.

3. The method according to claim 1, characterized in that: After the step of calculating the queue ratio and vehicle accumulation rate of each entrance road according to the real-time traffic data, and dividing the traffic state of each entrance road into a normal state, a warning state and a dangerous state according to a preset state threshold, the method further includes: When the traffic state is in the normal state, obtaining the vehicle clearance acceptability rate of the entrance road; When the vehicle gap acceptability rate is greater than a preset acceptance rate threshold, shortening the signal timing sequence by a preset time length; When the vehicle gap acceptability rate is not greater than the preset acceptance rate threshold, the signal timing sequence is extended by a preset time length.

4. The method according to claim 3, characterized in that After the step of extending the signal timing sequence for a preset time length when the vehicle gap acceptability rate is not greater than the preset acceptance rate threshold, the method further includes: Real-time monitoring of the changing trends of the queuing ratio and the vehicle accumulation rate; When it is detected that the queuing ratio continues to increase in a plurality of consecutive signal cycles and exceeds a first warning threshold, the traffic state is switched to the warning state.

5. The method according to claim 3, characterized in that: After the step of calculating the queue ratio and vehicle accumulation rate of each entrance road according to the real-time traffic data, and dividing the traffic state of each entrance road into a normal state, a warning state and a dangerous state according to a preset state threshold, the method further includes: When the traffic state is in the dangerous state, calculating the vehicle saturation of each of the entrance lanes, the vehicle saturation being equal to the product of the vehicle arrival rate and the vehicle gap acceptability rate; Prioritizing the entry lanes according to the vehicle saturation; Extend the green light period for the highest priority entrance lane to a preset time.

6. The method according to claim 5, characterized in that After the step of extending the green light period of the highest priority entrance lane to a preset time, the method further includes: When the queuing ratio decreases below a second warning threshold and the vehicle accumulation rate shows a downward trend, and the vehicle gap acceptance rate is within a preset normal range, the traffic state is switched to the warning state.

7. A second-level adaptive control system, characterized in that: The second-level adaptive control system includes: one or more processors and a memory; the memory is coupled to the one or more processors, the memory is used to store computer program code, the computer program code includes computer instructions, and the one or more processors call the computer instructions to enable the second-level adaptive control system to execute the method described in any one of claims 1-6.

8. A computer-readable storage medium comprising instructions, characterized in that: When the instruction runs on the second-level adaptive control system, the second-level adaptive control system executes the method as claimed in any one of claims 1 to 6.

9. A computer program product, characterized in that When the computer program product runs on a second-level adaptive control system, the second-level adaptive control system is enabled to execute the method according to any one of claims 1 to 6.

Images