A city-level timing strategy optimization method and system

Abstract

A city-level timing strategy optimization method and system. This method uses a modified Webster delay formula based on raw traffic data to calculate vehicle delays at each intersection in different directions. The traffic state at each intersection is divided into unimpeded, lightly congested, moderately congested, and heavily congested based on the vehicle delays at each intersection in different directions. A timing strategy optimization model is constructed based on the traffic state, with minimizing total delay time as the objective function. The signal cycle, phase duration, and phase difference at each intersection are used as timing parameters, and the optimal timing solution is obtained by iteratively calculating the solution of the timing strategy optimization model. The optimal timing solution is then distributed to the signal controller at each intersection. This application aims to resolve the problem of inconsistent timing plans between adjacent intersections, thereby improving road traffic efficiency.

Description

Technical Field

[0001] This application belongs to the field of intelligent transportation Internet of Things application services, and in particular relates to a city-level timing strategy optimization method and system. Background Art

[0002] With the continuous expansion of cities and the rapid growth of motor vehicle ownership, urban traffic congestion is becoming increasingly serious. Traditional traffic signal timing methods are mainly based on fixed time schemes or simple sensor-based control, which is difficult to adapt to the complex and changing urban traffic flow conditions. Especially during peak hours in the morning and evening, due to the significant increase in traffic volume, the lights are still operated according to the off-peak hours. This leads to long queues of vehicles on major roads, and they need to wait for multiple signal cycles to pass through the intersection. At night or during holidays, when traffic flow is low, the lights switch at regular time intervals, resulting in unnecessary idle cycles, wasting road resources and travelers' time.

[0003] In related technologies, geomagnetic sensors and video cameras are installed at intersections to collect traffic flow data. Traffic flow characteristics are analyzed and corresponding signal timing plans are developed. This method allows for a certain degree of adjustment of signal timing based on collected real-time traffic data, increasing the flexibility of traffic signal control.

[0004] However, since the above methods only focus on the traffic conditions of a single intersection or local area for independent timing optimization, they lack the ability to globally analyze and collaboratively optimize the traffic flow distribution patterns of the entire urban road network. As a result, the timing plans between adjacent intersections may be inconsistent, causing vehicles to frequently encounter red lights when passing through a series of intersections, reducing road traffic efficiency. Summary of the Invention

[0005] This application provides a city-level timing strategy optimization method and system for solving the problem of inconsistent timing plans between adjacent intersections, thereby improving road traffic efficiency.

[0006] First, the present application provides a city-level timing strategy optimization method that obtains traffic flow, speed, vehicle type, queue length, and lane occupancy at each intersection collected by multi-source traffic data collection equipment in real time to obtain raw traffic data;

[0007] Based on the original traffic data, the improved Webster delay formula is used to calculate the vehicle delay time in different directions at each intersection;

[0008] The traffic status of each intersection is divided into smooth, lightly congested, moderately congested and severely congested according to the vehicle delay time in different directions at each intersection;

[0009] Based on traffic conditions, a timing strategy optimization model is constructed with the minimum total delay time as the objective function. The urban road network is abstracted into a graph structure with each intersection as a node and each road section as an edge. The constraints of the timing strategy optimization model include minimum green light time constraint, maximum green light time constraint, and intersection saturation constraint.

[0010] The signal cycle, phase duration, and phase difference of each intersection are used as timing parameters, and the optimal timing solution is obtained by iteratively calculating the solution of the timing strategy optimization model.

[0011] The optimal timing plan is sent to the traffic light controllers at each intersection, so that all traffic light controllers operate according to the optimal timing plan.

[0012] By employing this technical solution, real-time multi-source traffic data is collected and delays are calculated based on a modified Webster delay formula, accurately reflecting the actual traffic conditions at each intersection. After classifying the traffic conditions at each intersection, a scheduling strategy optimization model is constructed with minimizing total delay as the objective function. This model abstracts the urban road network into a graph structure and considers constraints such as minimum green time, maximum green time, and intersection saturation, enabling the optimization process to comprehensively balance traffic demand at each intersection. By iteratively optimizing the signal cycle, phase duration, and phase difference as timing parameters, a scheduling solution is obtained that minimizes total network delay while satisfying various constraints. This scheduling solution can adapt to the dynamic changes in urban traffic flow, improve road network efficiency, reduce vehicle delays within the network, and enhance traffic quality.

[0013] In conjunction with some embodiments of the first aspect, in some embodiments, the vehicle delay time at each intersection in different directions is calculated using an improved Webster delay formula based on original traffic data, specifically including:

[0014] Preprocessing the original traffic data to obtain preprocessed traffic data, wherein the preprocessing includes eliminating abnormal data that exceeds a preset range;

[0015] Divide the pre-processed traffic data into time windows and calculate the average vehicle arrival rate and average service time of each intersection in each time window;

[0016] The average vehicle arrival rate and average service time are substituted into the improved Webster delay formula to calculate the vehicle delay time at each intersection in different directions.

[0017] By employing the above technical solution, preprocessing raw traffic data can eliminate outliers and improve data quality. Using a time window partitioning method to calculate average vehicle arrival rates and average service times can eliminate the impact of short-term random fluctuations. Substituting these preprocessed parameters into the improved Webster delay formula yields more accurate vehicle delay calculations. This result reflects the actual traffic conditions at each intersection in different directions, providing reliable delay data for subsequent traffic state classification and scheduling optimization, ensuring that the entire scheduling optimization process is based on accurate delay assessments.

[0018] In conjunction with some embodiments of the first aspect, in some embodiments, the traffic state of each intersection is divided into unimpeded, lightly congested, moderately congested, and severely congested according to the vehicle delay time at each intersection in different directions, specifically including:

[0019] The average delay time of each intersection in different time periods obtained by statistics is used as the benchmark threshold;

[0020] The benchmark threshold is used to calculate the deviation rate between the real-time delay time of each intersection and the benchmark threshold;

[0021] According to the deviation rate, a state evaluation index matrix is constructed, and the state evaluation index matrix is used to perform preliminary clustering to obtain the clustering results;

[0022] According to the clustering results, the traffic status of each intersection is divided into unimpeded, lightly congested, moderately congested and severely congested.

[0023] By employing this technical solution, historical data is collected to establish a baseline threshold, and a state evaluation index matrix is constructed using the deviation rate between real-time delay time and the baseline threshold. This provides objective evaluation criteria for traffic state classification. Cluster analysis based on this state evaluation index matrix allows intersections with similar traffic characteristics to be classified into corresponding traffic state levels, reducing the errors caused by subjective judgment. This makes the traffic state classification results more objective and accurate, truly reflecting the congestion level of each intersection and helping to accurately identify problem sections in the road network.

[0024] In conjunction with some embodiments of the first aspect, in some embodiments, a timing strategy optimization model with minimizing total delay time as an objective function is constructed based on traffic conditions, specifically including:

[0025] Calculate the capacity coefficient of each road section based on the traffic conditions of each intersection;

[0026] Establish a road network connectivity constraint matrix, the element values of which are determined according to the distance between adjacent intersections, the traffic arrival rate and the green wave speed;

[0027] Based on the capacity coefficient and the road network connectivity constraint matrix, the total delay time is expressed as a function of the signal timing parameters of each intersection. The function includes a random delay term and a uniform delay term. The random delay term is determined by the randomness of vehicle arrival, and the uniform delay term is determined by the signal timing plan.

[0028] Establish mathematical expressions for the minimum green time constraint, maximum green time constraint, and intersection saturation constraint. The minimum green time constraint is determined based on pedestrian crossing demand, the maximum green time constraint is determined based on vehicle queue length, and the intersection saturation constraint is determined based on traffic capacity.

[0029] The function is used as the objective function, and the minimum green light time constraint, maximum green light time constraint and intersection saturation constraint are used as constraints to obtain the timing strategy optimization model.

[0030] By employing this technical solution, the capacity coefficients of each road section were calculated and a network connectivity constraint matrix was established, quantifying the operational characteristics of each intersection and road section in the network. The total delay objective function constructed on this basis considers both random and uniform delay components, reflecting the impact of the signal timing scheme on vehicle delays. Furthermore, by introducing three constraints—minimum green time, maximum green time, and intersection saturation—the optimization model considers both pedestrian crossing needs and vehicle traffic efficiency, improving the feasibility and rationality of the timing scheme.

[0031] In conjunction with some embodiments of the first aspect, in some embodiments, after the optimal timing plan is sent to the traffic light controllers at each intersection, the method further includes:

[0032] A road network cycle fluctuation curve is constructed based on the historical timing data of each intersection. The road network cycle fluctuation curve reflects the changing pattern of the signal cycle of each intersection in the road network.

[0033] The local extreme point of the periodic fluctuation curve in the time dimension is taken as the key time node;

[0034] Calculate the cycle difference between two adjacent intersections at the key time node;

[0035] When the cycle difference is greater than a preset threshold, the road section between two adjacent intersections is marked as a potential bottleneck section;

[0036] Reverse timing optimization is performed on potential bottleneck sections. The reverse timing optimization is based on the upstream intersection and reversely adjusts the timing parameters of the downstream intersection to match the traffic capacity of the potential bottleneck section with the traffic volume.

[0037] By employing this technical solution, a road network cycle fluctuation curve is constructed, accurately reflecting the dynamic changes in signal cycles at each intersection in the road network. Combined with the calculation of cycle differences at key time nodes, potential traffic bottlenecks can be promptly identified. Reverse timing optimization is performed on identified potential bottlenecks, adjusting the timing parameters of downstream intersections inversely based on upstream intersections. This ensures that the capacity of potential bottlenecks matches actual traffic volume, preventing the occurrence and spread of traffic congestion and improving the overall efficiency of the road network.

[0038] In conjunction with some embodiments of the first aspect, in some embodiments, constructing a road network periodic fluctuation curve based on historical timing data of each intersection specifically includes:

[0039] The sliding time window method is used to segment the historical timing data of each intersection;

[0040] Wavelet transform is used to perform multi-scale decomposition on the data in each time window to extract the main frequency components of periodic changes;

[0041] The periodic variation trend is reconstructed based on the main frequency components to generate a smooth periodic fluctuation curve of the road network.

[0042] By adopting the above technical solution, the sliding time window method is used to segment the historical timing data, and combined with the multi-scale decomposition technology of wavelet transform, the main frequency components of periodic changes can be effectively extracted, and a smooth road network periodic fluctuation curve can be generated through reconstruction. Random fluctuations and noise interference in the timing data can be filtered out, and the true periodic change characteristics are retained. The extraction accuracy of periodic fluctuation characteristics is improved, making the subsequent bottleneck section identification based on the curve more accurate and reliable, providing a more precise decision-making basis for road network timing optimization.

[0043] In conjunction with some embodiments of the first aspect, in some embodiments, calculating the cycle difference between two adjacent intersections at a key time node specifically includes:

[0044] Establish intersection signal timing feature vectors, including intersection signal timing feature vector cycle length, green-to-signal ratio, phase difference, and flow direction distribution;

[0045] The feature vector distance between the timing feature vectors of two adjacent intersections is calculated using a multi-dimensional feature distance calculation method based on improved cosine similarity.

[0046] Calculate the road section saturation weight coefficient based on the road section capacity and actual demand;

[0047] Multiply the eigenvector distance by the saturation weight coefficient to obtain the period difference.

[0048] By adopting this technical solution, multiple key parameters influencing intersection timing coordination are comprehensively considered and uniformly expressed and processed in the form of feature vectors. The improved cosine similarity calculation method reduces the limitations of traditional Euclidean distance in processing high-dimensional features, and can more accurately measure the similarity of timing plans. The introduction of a saturation weighting factor accounts for the impact of actual traffic demand on the timing difference assessment, making the difference calculation results more consistent with actual traffic operation characteristics.

[0049] In the second aspect, an embodiment of the present application provides a city-level timing strategy optimization system, which 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 one or more processors call the computer instructions to enable the system to execute the method described in the first aspect and any possible implementation method of the first aspect.

[0050] In a third aspect, an embodiment of the present application provides a computer-readable storage medium comprising instructions, which, when executed on a system, enables the system to execute the method described in the first aspect and any possible implementation of the first aspect.

[0051] In a fourth aspect, an embodiment of the present application provides a computer program product, which, when executed on a system, enables the system to execute the method described in any possible implementation manner in the first aspect.

[0052] One or more technical solutions provided in the embodiments of this application have at least the following technical effects or advantages:

[0053] 1. This application provides a city-level timing strategy optimization method, which can accurately reflect the actual traffic operation status of each intersection by acquiring multi-source traffic data in real time and calculating the delay time based on the improved Webster delay formula. After the traffic status of each intersection is divided into levels, a timing strategy optimization model with the minimum total delay time as the objective function is constructed. The model abstracts the urban road network into a graph structure and takes into account constraints such as the minimum green light time, the maximum green light time and the intersection saturation, so that the optimization process can comprehensively weigh the traffic needs of each intersection. By iteratively optimizing the signal light cycle, phase duration and phase difference as timing parameters, a timing plan that minimizes the total network delay time while meeting various constraints can be obtained. This timing plan can adapt to the dynamic changes of urban traffic flow, improve the traffic efficiency of the road network, reduce the delay time of vehicles in the road network, and improve the quality of traffic operation.

[0054] 2. This application provides a city-level timing strategy optimization method, which calculates the capacity coefficient of each road section and establishes a road network connectivity constraint matrix, and quantifies the operating characteristics of each intersection and road section in the road network. The total delay time objective function constructed on this basis takes into account the two components of random delay and uniform delay, reflecting the impact of the signal timing plan on vehicle delays. At the same time, the three types of constraints, namely minimum green light time, maximum green light time and intersection saturation, are introduced, so that the optimization model takes into account both the needs of pedestrians crossing the street and the efficiency of vehicle traffic, thereby improving the feasibility and rationality of the timing plan.

[0055] 3. This application provides a city-level timing strategy optimization method that constructs a road network cycle fluctuation curve that can accurately reflect the dynamic changes in the signal cycle of each intersection in the road network. Combined with the cycle difference calculation of key time nodes, potential traffic bottleneck sections can be identified in a timely manner. Reverse timing optimization is performed on the identified potential bottleneck sections. The timing parameters of the downstream intersection are adjusted inversely based on the upstream intersection. This ensures that the traffic capacity of the potential bottleneck section matches the actual traffic volume. This can prevent the occurrence and spread of traffic congestion and improve the overall traffic efficiency of the road network. BRIEF DESCRIPTION OF THE DRAWINGS

[0056] Figure 1 It is a flow chart of a city-level timing strategy optimization method in an embodiment of the present application.

[0057] Figure 2 This is another flow chart of a city-level timing strategy optimization method in an embodiment of the present application.

[0058] Figure 3 This is a schematic diagram of the physical device structure of a city-level timing strategy optimization system provided in an embodiment of the present application. DETAILED DESCRIPTION

[0059] The terms used in the following examples 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 and appended claims of this application, the singular expressions "a," "an," "said," "above," "the," and "this" are intended to include plural expressions as well, unless the context clearly indicates otherwise. It should also be understood that the term "and / or" used in this application refers to any or all possible combinations comprising one or more of the listed items.

[0060] In the following, the terms "first" and "second" are used for descriptive purposes only and should not be understood to imply or suggest relative importance or implicitly indicate the number of the technical features indicated. Therefore, the features defined as "first" and "second" may explicitly or implicitly include one or more of the features. In the description of the embodiments of this application, unless otherwise specified, "plurality" means two or more.

[0061] The following uses an embodiment and combines Figure 1 , a city-level timing strategy optimization method in an embodiment of the present application is described:

[0062] See also Figure 1 , which is a flow chart of a city-level timing strategy optimization method in an embodiment of the present application.

[0063] S101, acquiring in real time the traffic flow, vehicle speed, vehicle type, queue length, and lane occupancy of each intersection collected by multi-source traffic data collection equipment to obtain raw traffic data;

[0064] The system requires real-time traffic data from every intersection in the city as input for subsequent optimization. Traffic data can come from a variety of collection devices, such as geomagnetic detectors, video surveillance, and microwave vehicle detectors. These devices, located at every intersection in the city, can collect real-time data such as traffic flow, average vehicle speed, vehicle type composition, queue length, and lane occupancy. The system aggregates data from these collection devices to generate raw traffic data for the entire city.

[0065] In practical implementation, the system can directly interface with these data collection devices, receiving data uploaded by them in real time via wired or wireless communication. To ensure data timeliness, the system can require data collection devices to report data at a frequency of seconds or minutes. Furthermore, the system can set up data caching to prevent data loss due to communication interruptions and other reasons. The obtained raw data can be stored in a structured form (such as a relational database) or an unstructured form (such as a text file) to prepare for subsequent computational analysis.

[0066] S102. Calculate the vehicle delay time at each intersection in different directions using the improved Webster delay formula based on the original traffic data;

[0067] Based on the original traffic data, the system uses the improved Webster delay formula to calculate the vehicle delay time in different directions at each intersection. Specifically, the original traffic data is preprocessed to obtain preprocessed traffic data. The preprocessing includes eliminating abnormal data that exceeds the preset range;

[0068] Divide the pre-processed traffic data into time windows and calculate the average vehicle arrival rate and average service time of each intersection in each time window;

[0069] The average vehicle arrival rate and average service time are substituted into the improved Webster delay formula to calculate the vehicle delay time at each intersection in different directions.

[0070] After obtaining raw traffic data from each intersection, the system needs to evaluate traffic conditions in different directions at each intersection. Vehicle delay is a commonly used evaluation metric, reflecting the additional time a vehicle consumes due to interference from traffic lights and other vehicles while passing through the intersection. The traditional Webster delay formula takes into account parameters such as vehicle arrival rate, signal cycle, and green light duration, but it lacks a detailed understanding of actual road conditions. This step utilizes a modified Webster delay formula to more accurately estimate delay time.

[0071] Specifically, the system first pre-processes the raw traffic data, removing data points with obvious anomalies to ensure data reliability. The data is then divided into fixed time windows (e.g., 5 minutes) to calculate the average vehicle arrival rate and average service time (i.e., the time it takes to pass through the intersection) for each direction at each intersection within each time window. These two parameters are key factors affecting delays. Finally, the average arrival rate and service time are substituted into the improved Webster delay formula to estimate the delay time at each direction at each intersection during each time period. Compared to the classic formula, the improved formula may incorporate more parameters that reflect the actual road conditions, such as the number of lanes, turn ratio, pedestrian crossing time, etc., thereby obtaining a delay estimate that is closer to reality.

[0072] S103, classifying the traffic status of each intersection into unimpeded, lightly congested, moderately congested, and severely congested according to the vehicle delay time at each intersection in different directions;

[0073] The system divides the traffic status of each intersection into smooth, lightly congested, moderately congested, and heavily congested based on the vehicle delay time at each intersection in different directions. Specifically, the average delay time of each intersection in different time periods obtained by statistics is used as the benchmark threshold;

[0074] The benchmark threshold is used to calculate the deviation rate between the real-time delay time of each intersection and the benchmark threshold;

[0075] According to the deviation rate, a state evaluation index matrix is constructed, and the state evaluation index matrix is used to perform preliminary clustering to obtain the clustering results;

[0076] According to the clustering results, the traffic status of each intersection is divided into unimpeded, lightly congested, moderately congested and severely congested.

[0077] After calculating vehicle delays in all directions at each intersection, the system needs to further assess traffic conditions at each intersection to inform subsequent signal timing optimization decisions. This step categorizes traffic conditions at intersections into four levels: smooth, lightly congested, moderately congested, and severely congested, based on the magnitude of the delays. Smooth traffic indicates minimal traffic obstruction, lightly congested traffic indicates slight delays but generally smooth traffic flow, and moderately and severely congested traffic indicates significant traffic disruption and urgent need for traffic diversion.

[0078] In implementation, the system first statistically analyzes historical data to determine the average delay time at each intersection over different time periods, using this as a benchmark threshold for traffic status classification. For example, a delay time 20% below the historical average could be considered unimpeded, a delay time 20% to 50% above the average could be considered mild congestion, a delay time 50% to 100% above the average could be considered moderate congestion, and a delay time above 100% could be considered severe congestion. During real-time operation, the system calculates the real-time delay time at each intersection and compares it with the benchmark threshold to determine the current traffic status level for that intersection.

[0079] Given the complexity of actual road conditions, it can be difficult to accurately determine traffic status based solely on the delay time at a single moment. To this end, the system can combine other indicators to adopt a more refined state classification method. For example, the changing trend of delay time over a period of time can be calculated. If delays continue to increase, it indicates that congestion is worsening. Another example is the average speed and acceleration of queued vehicles. The lower the speed and the more frequent idling, the higher the congestion. By aggregating these indicators into a state evaluation matrix and then performing cluster analysis on each intersection, more accurate and comprehensive state judgment results can be obtained.

[0080] It's important to note that road congestion evolves dynamically, influenced by factors such as upstream and downstream road sections, weather conditions, and accident handling. Therefore, while determining the status of an intersection, the system also analyzes the causes of this evolution and estimates its duration and development trends. This facilitates decision-making optimization and resource scheduling. For example, if the system predicts that a certain intersection will transition from mild to moderate congestion within the next half hour, it can adjust the signal timing at the relevant intersection in advance, diverting traffic flow and preventing further congestion. Conversely, if the congestion is caused by a temporary event, the system can coordinate emergency response to shorten the spatial and temporal scope of the event's impact.

[0081] S104. Constructing a timing strategy optimization model based on traffic conditions with minimizing total delay time as the objective function;

[0082] Based on traffic conditions, the system constructs a timing strategy optimization model with minimizing total delay time as the objective function. The urban road network is abstracted into a graph structure, with each intersection as a node and each road section as an edge. The constraints of the timing strategy optimization model include minimum green light time constraint, maximum green light time constraint, and intersection saturation constraint. Specifically, the capacity coefficient of each road section is calculated based on the traffic conditions of each intersection.

[0083] Establish a road network connectivity constraint matrix, the element values of which are determined according to the distance between adjacent intersections, the traffic arrival rate and the green wave speed;

[0084] Based on the capacity coefficient and the road network connectivity constraint matrix, the total delay time is expressed as a function of the signal timing parameters of each intersection. The function includes a random delay term and a uniform delay term. The random delay term is determined by the randomness of vehicle arrival, and the uniform delay term is determined by the signal timing plan.

[0085] Establish mathematical expressions for the minimum green time constraint, maximum green time constraint, and intersection saturation constraint. The minimum green time constraint is determined based on pedestrian crossing demand, the maximum green time constraint is determined based on vehicle queue length, and the intersection saturation constraint is determined based on traffic capacity.

[0086] The function is used as the objective function, and the minimum green light time constraint, maximum green light time constraint and intersection saturation constraint are used as constraints to obtain the timing strategy optimization model.

[0087] After obtaining the traffic state classification results for each intersection, the system needs to build a signal timing optimization model based on this data to solve the optimal timing solution. The model's goal is to minimize the total delay time for vehicles and pedestrians in the area while ensuring safety and fairness. This step describes the model construction process in detail.

[0088] First, the system abstracts the entire city road network into a graph model. Intersections correspond to nodes in the graph, and roads correspond to the edges connecting the nodes. Each node and edge has a series of attributes, such as the node's traffic status level and traffic flow, and the edge length, speed limit, and capacity. These attributes are mostly derived from the calculation results of the previous steps, with a small number of parameters (such as road length) coming from the underlying map data. Converting the road network into a graph model facilitates the implementation of subsequent optimization algorithms.

[0089] Next, the system needs to establish the objective function of the optimization model—a mathematical expression for the total delay time. Total delay is divided into two components: random delay and uniform delay. Random delay is primarily caused by the randomness of vehicle arrivals and is related to traffic flow at the intersection; uniform delay results from signal stop times and is directly related to timing parameters. When calculating delay, the system fully considers the actual road capacity. For example, the capacity between adjacent intersections depends not only on the length of the road segment but also on factors such as the saturated traffic volume of the segment, the vehicle arrival rate, and the speed of the green wave zone under coordinated control. The weighted average of the capacity of each segment serves as the capacity coefficient for the entire road network. Furthermore, the system defines a network connectivity matrix, in which each element indicates whether green wave coordination can be achieved between two adjacent intersections. Green wave coordination allows vehicles to continuously pass through multiple intersections at a specific speed without being blocked by traffic lights. However, not all road segments meet green wave conditions such as length and speed. Incorporating the capacity coefficient and connectivity matrix into the delay function yields a more realistic delay estimate.

[0090] Finally, the system adds various constraints based on the objective function to build a complete timing optimization model. Common constraints include:

[0091] Minimum and maximum green light duration constraints. To ensure the right of way for pedestrians and vehicles in all directions, the green light duration for each phase must not fall below a certain lower limit. To prevent excessive green light durations that cause long waits for other directions, the green light duration must not exceed a certain upper limit. The specific values of these upper and lower limits can be adjusted dynamically based on pedestrian crossing demand and vehicle flow conditions at each intersection.

[0092] Intersection saturation constraint. Saturation represents the ratio of the number of vehicles passing through an intersection in a cycle to its capacity. When saturation exceeds 1, the number of vehicles arriving exceeds the intersection's capacity, leading to an indefinite backlog and the need to extend the green light duration. Therefore, the saturation of each intersection should not exceed 1, serving as a hard constraint on the timing plan.

[0093] In addition to the common constraints mentioned above, other constraints may be encountered in practical applications, such as the range of signal differences between adjacent intersections, and the need for priority for buses and emergency vehicles. These constraints can be incorporated into the model as appropriate and assigned appropriate weights. This completes the construction of a multi-constrained, nonlinear mixed integer programming model. This model encompasses the key characteristics of road network traffic and can effectively guide the optimization of scheduling solutions.

[0094] S105, using the signal cycle, phase duration, and phase difference of each intersection as timing parameters, and iteratively calculating the solution of the timing strategy optimization model to obtain the optimal timing solution;

[0095] After constructing the timing optimization model, the system uses an optimization algorithm to find the optimal solution to the model, i.e., the optimal timing plan. This step uses the key parameters involved in the timing plan (traffic light cycle, phase duration, and phase difference) as decision variables and uses an iterative calculation method to search for the optimal parameter combination, thereby obtaining the globally optimal timing strategy.

[0096] In practical implementation, the system generally employs heuristic algorithms or intelligent optimization algorithms to solve the model. These algorithms repeatedly iteratively approach the optimal solution, achieving a satisfactory solution within an acceptable timeframe while also adapting to the model's nonlinear and multi-constrained characteristics. For example, a genetic algorithm simulates the process of biological evolution, performing selection, crossover, and mutation on a set of randomly generated initial solutions (i.e., timing parameter combinations), continuously generating new solutions with improved performance and ultimately converging on the optimal solution. Similarly, an ant colony algorithm simulates the food-seeking behavior of ants, using the accumulation and volatilization of pheromones to guide the search path toward the global optimal solution. The key to these algorithms lies in balancing solution quality and convergence speed. The former requires extensive exploration within the search space, while the latter requires rapid identification of promising areas. By adjusting algorithmic parameters such as population size, number of iterations, crossover probability, and pheromone concentration, an appropriate trade-off can be found between these two objectives.

[0097] In the search for the optimal solution, the relevant parameters of traffic lights at each intersection are simultaneously considered as decision variables. Typically, signal timing at adjacent intersections within a region needs to be coordinated to achieve green wave traffic flow and reduce parking queues. Therefore, optimization should not be conducted on a discrete basis, but rather on the road network as a whole. While computationally intensive, this globally coordinated timing approach fully exploits the potential capacity of the road network and offers advantages over local optimization methods such as single-point control or coordinated control of primary and secondary roads.

[0098] The urban road network is a dynamically changing system, and the calculated optimal timing solution can only remain optimal for a limited time. To adapt to new traffic demands, the system needs to periodically re-execute the optimization process and adjust the timing parameters. Therefore, a fast and efficient solution algorithm is crucial for optimal timing. In addition to selecting high-performance intelligent algorithms, the system can also partition the road network and perform parallel calculations on each zone; discretize the timing parameters to reduce the solution space; and implement pruning during the calculation process to eliminate obvious inferior solutions as early as possible. These measures all help to accelerate the search for the optimal solution.

[0099] S106: Send the optimal timing plan to the traffic light controllers at each intersection, so that all traffic light controllers operate according to the optimal timing plan.

[0100] After successfully solving the optimal timing solution, the remaining work is to apply it to actual traffic light control, aiming to achieve coordinated and efficient signal timing across the entire road network and improve traffic efficiency. This step describes the process of issuing and implementing the optimal solution.

[0101] Specifically, the system first converts the optimal timing plan into a series of control instructions, mainly including parameters such as the cycle duration, phase sequence, and green light duration of each intersection. These instructions are generally encapsulated in standardized data formats, such as XML or JSON, to ensure that the information is complete and accurate during transmission and parsing. Then, the system sends the control instructions to the traffic light controllers at the corresponding intersections through wired or wireless communication networks (such as GPRS, 4G, NB-IoT, etc.). The controller is an embedded device deployed at the intersection, which directly controls the display of traffic lights through the I / O interface. It obtains timing instructions from the system, switches traffic lights strictly in accordance with the specified duration, phase sequence and other parameters, and finally implements the timing plan.

[0102] In the above-described embodiment, by acquiring multi-source traffic data in real time and calculating delays based on the improved Webster delay formula, the actual traffic status at each intersection can be accurately reflected. After classifying the traffic status at each intersection, a timing strategy optimization model is constructed with minimizing total delay as the objective function. This model abstracts the urban road network into a graph structure and considers constraints such as minimum green time, maximum green time, and intersection saturation, enabling the optimization process to comprehensively balance traffic demand at each intersection. By iteratively optimizing signal cycle, phase duration, and phase difference as timing parameters, a timing solution can be obtained that minimizes total network delay while satisfying various constraints. This timing solution can adapt to the dynamic changes in urban traffic flow, improve road network efficiency, reduce vehicle delays within the network, and enhance traffic quality.

[0103] In order to further improve the reliability and practicality of the above-mentioned timing strategy optimization method, the embodiment of the present application also provides a bottleneck road section identification and optimization method based on historical timing data. This method analyzes the historical data of signal timing at each intersection in the road network, identifies the road sections that are prone to traffic bottlenecks, and uses reverse optimization to adjust the timing plan in advance, thereby preventing the occurrence of traffic congestion. Figure 2 , a bottleneck section identification and optimization method based on historical timing data in an embodiment of the present application is described:

[0104] See also Figure 2 , which is a flow chart of a bottleneck section identification and optimization method based on historical timing data in an embodiment of the present application.

[0105] S201, constructing a road network periodic fluctuation curve based on historical timing data of each intersection;

[0106] The system constructs a road network cycle fluctuation curve based on the historical timing data of each intersection. The road network cycle fluctuation curve reflects the changing pattern of the signal cycle at each intersection in the road network, specifically including:

[0107] The sliding time window method is used to segment the historical timing data of each intersection;

[0108] Wavelet transform is used to perform multi-scale decomposition on the data in each time window to extract the main frequency components of periodic changes;

[0109] The periodic variation trend is reconstructed based on the main frequency components to generate a smooth periodic fluctuation curve of the road network.

[0110] This step analyzes historical data on signal timing at each intersection in the road network to identify long-term patterns in signal cycles, laying the foundation for identifying traffic bottlenecks and optimizing timing plans. Because real-world road traffic exhibits distinct temporal characteristics, such as peak hours and weekday / weekend variations, signal cycles often exhibit cyclical patterns. Exploiting these cyclical patterns can better characterize the dynamics of road network traffic, predict future traffic flow trends, and ultimately implement targeted signal timing strategies.

[0111] In its specific implementation, the system first uses a sliding time window method to segment historical timing data. The sliding window divides the continuous time series data into several overlapping time segments by setting a fixed time span (such as 1 hour) and a sliding step size (such as 15 minutes), each segment corresponding to an observation window. The system then performs a multi-scale decomposition on the data within each window. Because traffic cycles have different time granularities (such as hours, days, weeks, etc.), traditional Fourier transforms are difficult to fully extract these multi-scale features. Therefore, the system uses a wavelet transform method to project the window data onto different time-frequency scales by selecting different wavelet bases, obtaining a series of wavelet coefficients that reflect the intensity of periodic changes at different scales. Finally, the system selects wavelet coefficients with larger energy to reconstruct the main change trend of the signal period within the window, and connects the change trends of each window in chronological order to form a smooth and continuous road network periodic fluctuation curve.

[0112] S202, taking the local extreme point of the periodic fluctuation curve in the time dimension as the key time node;

[0113] After successfully constructing the road network cycle fluctuation curve, the system needs to further determine the time nodes that have a significant impact on traffic conditions. This step uses the local extreme points of the fluctuation curve (i.e., local maximum and minimum points) as key nodes. These nodes often correspond to moments of sudden changes in traffic conditions, such as the beginning and end of morning and evening rush hours, the occurrence and recovery of traffic accidents, etc. At these moments, the contradiction between road capacity and traffic volume is most prominent, and the slightest carelessness may cause traffic bottlenecks. Therefore, it is necessary to focus on the coordination of the timing of the intersections before and after these nodes to predict the risk of traffic bottlenecks.

[0114] In practice, the system applies numerical analysis to the road network's periodic fluctuation curves, calculating first- and second-order derivatives. The sign changes of the derivatives reflect the monotonicity of the curve. Points where the first-order derivative is zero and the second-order derivative is less than zero are local maxima, while points where the first-order derivative is zero and the second-order derivative is greater than zero are local minima. For these extreme points, the system records their corresponding timestamps, which serve as key time points for subsequent analysis.

[0115] To ensure that the obtained extreme points truly reflect sudden changes in traffic conditions, the system can appropriately smooth the curve to eliminate extreme points caused by noise and subtle disturbances. For example, an amplitude threshold is set for traffic flow fluctuations based on the road grade. When the amplitude change between two adjacent extreme points is less than the threshold, they are merged into the same extreme interval, and the point with the largest amplitude change within the interval is selected as the key node. In addition, the system can also combine expert experience to manually set key nodes with universal patterns, such as statutory holidays and major event periods. Although such nodes may not be obvious in historical data, they still have reference value for traffic prediction in the future.

[0116] S203, calculating the cycle difference between two adjacent intersections at a key time node;

[0117] The system calculates the cycle difference between two adjacent intersections at key time nodes, specifically by: establishing the intersection signal timing feature vector, including the intersection signal timing feature vector cycle length, green-to-signal ratio, phase difference, and flow direction distribution;

[0118] The feature vector distance between the timing feature vectors of two adjacent intersections is calculated using a multi-dimensional feature distance calculation method based on improved cosine similarity.

[0119] Calculate the road section saturation weight coefficient based on the road section capacity and actual demand;

[0120] Multiply the eigenvector distance by the saturation weight coefficient to obtain the period difference.

[0121] After identifying critical time nodes, the system needs to assess the timing differences between adjacent intersections at those nodes to predict the risk of traffic bottlenecks. This step introduces the concept of periodic discrepancy, which quantifies the spatial deviation in the timing strategies of adjacent intersections at critical nodes. The greater the discrepancy, the more inconsistent the timing is between adjacent intersections, and the greater the potential for traffic bottlenecks. Calculating this discrepancy provides an intuitive and measurable basis for subsequent bottleneck identification.

[0122] In the specific implementation, the system first establishes the characteristic vector of intersection signal timing. The characteristic vector contains multiple key parameters that affect timing, such as signal cycle length, green-to-signal ratio (i.e., the proportion of green light duration), phase difference (i.e., the time difference between the start time of green lights at adjacent intersections), and traffic flow distribution. Then, the system uses the improved cosine similarity method to calculate the multidimensional distance between characteristic vectors. The improved cosine similarity not only considers the size of the features of each dimension, but also considers the correlation between different feature dimensions. For example, the cycle length is usually negatively correlated with the green-to-signal ratio, while the phase difference is positively correlated with the flow distribution. When calculating the distance, these features should be given different weights. Finally, the similarity measure is normalized to the interval [0, 1], and the resulting value is the cycle difference. The closer the value is to 1, the greater the difference in timing strategies.

[0123] When calculating variance, the importance of adjacent intersections is typically not the same. For intersections located on main roads with high traffic volume, timing coordination is more effective in alleviating bottlenecks. Therefore, the system also needs to calculate segment weights to correct variance. Calculating segment weights requires comprehensive consideration of multiple factors, such as the segment's capacity, saturation, speed, and queue length. Capacity reflects the physical properties of a segment, such as the number of lanes and lane width; saturation reflects the level of traffic flow; higher values indicate more frequent conflicts; speed reflects the driving conditions; below a threshold, delays may occur; and queue length reflects the segment's reserve capacity; exceeding a threshold indicates overflow risk. The system then weights and sums these factors to obtain a comprehensive weight for the segment, which is then multiplied by the original variance to produce the corrected period variance.

[0124] S204: When the cycle difference is greater than a preset threshold, the road section between two adjacent intersections is marked as a potential bottleneck section;

[0125] Based on periodic differences, the system can initially identify potential bottleneck sections in the road network. This step sets a difference threshold. When the difference between adjacent intersections exceeds the threshold, the section connecting the two intersections is determined to be at risk of forming a traffic bottleneck and is subsequently marked as a potential bottleneck section. The threshold setting must balance accuracy and comprehensiveness of identification while fully considering the actual traffic conditions of the road network. Setting the threshold too high may miss some bottleneck sections; setting it too low may over-mark sections with less traffic pressure, affecting the targeted nature of subsequent optimization.

[0126] During specific implementation, the system can use a hierarchical threshold method to set different difference thresholds for roads of different levels. For example, the threshold for expressways and main roads is set to 0.6, the threshold for secondary roads is set to 0.7, and the threshold for branch roads is set to 0.8. When the difference between adjacent intersections exceeds the threshold corresponding to the level of the road they are on, the road section is marked as a potential bottleneck. The setting of hierarchical thresholds should be based on long-term traffic observation data and expert experience. It should be adapted to the traffic capacity of the road section and take into account the distribution characteristics of bottlenecks in the entire road network. In actual applications, an adaptive threshold method can also be used to dynamically adjust the threshold according to the overall traffic conditions of the road network. For example, the threshold can be appropriately lowered during rush hours in the morning and evening to increase the sensitivity of bottleneck identification; the threshold can be appropriately increased at night when traffic is sparse to reduce the misjudgment rate.

[0127] In addition, due to random disturbances in road network topology and traffic flow, isolated bottleneck sections may not truly reflect traffic conditions and may interfere with optimization decisions. Therefore, when identifying bottleneck sections, the system also needs to consider spatiotemporal correlation characteristics. For example, the mean difference between multiple consecutive sections on the same road can be calculated. When the mean exceeds a threshold, these sections are marked as potential bottlenecks. For example, if a section is frequently marked as a bottleneck over a period of time, the weight of its bottleneck possibility should be increased. By fusing spatiotemporal features, the reliability of identifying potential bottleneck areas can be further improved.

[0128] S205: Optimize the reverse timing of potential bottleneck sections.

[0129] The system performs reverse timing optimization on potential bottleneck sections. The reverse timing optimization uses the upstream intersection as a benchmark and reversely adjusts the timing parameters of the downstream intersection to match the traffic capacity of the potential bottleneck section with the traffic volume.

[0130] After identifying potential bottleneck sections, the system needs to optimize the timing plans for these sections in a targeted manner to alleviate or avoid traffic bottlenecks. Traditional timing optimization mostly adopts the idea of forward adjustment, that is, starting from the starting intersection, adjusting the timing of each intersection one by one along the direction of traffic flow, so that it can adapt to upstream vehicles as much as possible. Although this idea is intuitive, it has certain limitations: when the traffic capacity of the downstream section cannot meet the traffic demand of the upstream, no matter how the timing is adjusted, it cannot fundamentally alleviate the bottleneck. Therefore, this step proposes a new idea of reverse timing optimization, starting from the downstream intersection of the bottleneck section, and adjusting the timing of the upstream intersection in reverse to make it coordinated with the traffic capacity of the downstream section. This method is equivalent to treating the bottleneck source, which is more conducive to fundamentally alleviating congestion.

[0131] In implementation, the system first identifies key intersections downstream of potential bottlenecks (such as intersections and entrances / exits) and assesses their capacity, including the number of vehicles available and the number of vehicles in queue. Then, based on actual traffic flow demand, the system sets an initial timing plan at the key intersection, with key parameters including capacity, saturation, and delay time. Next, starting from the downstream intersection, the system reversely adjusts the timing parameters of each upstream intersection, primarily including cycle length, green-to-signal ratio, and phase difference. The goal of this adjustment is to ensure that traffic demand and capacity between upstream and downstream intersections are closely aligned, with the cycle difference below a threshold. This adjustment process must adhere to various constraints, such as cycle length range, minimum green time, and pedestrian crossing requirements. Furthermore, the timing plans of adjacent roads must be coordinated to prevent local adjustments from causing global fluctuations. Optimization algorithms such as genetic algorithms and particle swarm optimization can be used to search for the optimal timing parameter combination within these constraints.

[0132] The key to reverse scheduling optimization is estimating the capacity of a road section. Because real-world roads often experience oversaturation, the capacity of a road section depends not only on its physical properties but also on upstream and downstream traffic conditions. Therefore, when evaluating capacity, the system can employ dynamic estimation methods, taking into account factors such as traffic volume, queue length, and delay time on upstream and downstream sections in real time. For example, when traffic volume on a downstream section is high, the estimated capacity of the current section should be appropriately lowered; when there are many vehicles queuing on an upstream section, the capacity should also be adjusted downward to reduce the pressure on the upstream queue. Combining dynamic capacity with real-time traffic parameters can more accurately characterize the capacity constraints of a road section, resulting in more robust scheduling optimization results.

[0133] In the above-described embodiment, by constructing a road network cycle fluctuation curve, the dynamic changes in signal cycles at each intersection in the road network can be accurately reflected. Combined with the calculation of cycle differences at key time nodes, potential traffic bottlenecks can be promptly identified. Reverse timing optimization is performed on identified potential bottlenecks, adjusting the timing parameters of downstream intersections inversely based on upstream intersections. This ensures that the capacity of the potential bottleneck section matches the actual traffic volume, preventing the occurrence and spread of traffic congestion and improving the overall traffic efficiency of the road network.

[0134] The following describes the system in the embodiment of the present invention from the perspective of hardware processing. Figure 3 , which is a schematic diagram of the physical device structure of a city-level timing strategy optimization system provided in an embodiment of the present application.

[0135] It should be noted that Figure 3 The structure of the system shown is only an example and should not limit the functions and scope of use of the embodiments of the present invention.

[0136] like Figure 3As shown, the system includes a central processing unit (CPU) 301, which can perform various appropriate actions and processes, such as the methods described in the above embodiments, based on programs stored in a read-only memory (ROM) 302 or programs loaded from a storage unit 308 into a random access memory (RAM) 303. RAM 303 also stores various programs and data required for system operation. CPU 301, ROM 302, and RAM 303 are interconnected via a bus 304. An input / output (I / O) interface 305 is also connected to bus 304.

[0137] The following components are connected to the I / O interface 305: an input section 306 including a camera, infrared sensor, and the like; an output section 307 including a liquid crystal display (LCD) and speakers; a storage section 308 including a hard disk and the like; and a communication section 309 including a network interface card such as a LAN (Local Area Network) card or a modem. The communication section 309 performs communication processing via a network such as the Internet. A drive 310 is also connected to the I / O interface 305 as needed. Removable media 311, such as a magnetic disk, optical disk, magneto-optical disk, or semiconductor memory, is installed in the drive 310 as needed, so that computer programs read from the media can be installed in the storage section 308 as needed.

[0138] In particular, according to embodiments of the present invention, the processes described above with reference to the flowcharts can be implemented as computer software programs. For example, embodiments of the present invention include a computer program product comprising a computer program carried on a computer-readable medium, the computer program including a computer program for executing the methods illustrated in the flowcharts. In such embodiments, the computer program can be downloaded and installed from a network via the communication section 309 and / or installed from removable media 311. When executed by the central processing unit (CPU) 301, the computer program performs the various functions defined in the present invention.

[0139] It should be noted that the computer-readable medium described in the embodiments of the present invention may be a computer-readable signal medium or a computer-readable storage medium, or any combination thereof. A computer-readable storage medium may be, for example, but not limited to, an electrical, magnetic, optical, electromagnetic, infrared, or semiconductor system, apparatus, or device, or any combination thereof. More specific examples of computer-readable storage media may include, but are not limited to, an electrical connection having one or more conductors, a portable computer disk, a hard disk, random access memory (RAM), read-only memory (ROM), erasable programmable read-only memory (EPROM), flash memory, optical fiber, a portable compact disc read-only memory (CD-ROM), an optical storage device, a magnetic storage device, or any suitable combination thereof. In the present invention, a computer-readable storage medium may be any tangible medium containing or storing a program that can be used by or in conjunction with an instruction execution system, apparatus, or device. In the present invention, a computer-readable signal medium may include a data signal transmitted in baseband or as part of a carrier wave, which carries a computer-readable computer program. Such a propagated data signal may take any of a variety of forms, including but not limited to electromagnetic signals, optical signals, or any suitable combination thereof.

[0140] The flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions and operations of the systems, methods and computer program products according to various embodiments of the present invention. Each box in the flowchart or block diagram can represent a module, program segment, or part of the code, and the above-mentioned module, program segment, or part of the 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 boxes can also occur in an order different from that marked in the accompanying drawings. For example, two boxes shown in succession can actually be executed substantially in parallel, and they can sometimes be executed in the opposite order, depending on the functions involved. It should also be noted that each box in the block diagram or flowchart, and the combination of boxes in the block diagram or flowchart, can be implemented using a dedicated hardware-based system that performs the specified function or operation, or can be implemented using a combination of dedicated hardware and computer instructions.

[0141] As another aspect, the present invention further provides a computer-readable storage medium, which may be included in the system described in the above embodiments, or may exist independently and not incorporated into the system. The storage medium carries one or more computer programs, and when executed by a processor of a system, the system implements the methods provided in the above embodiments.

[0142] As described 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 above embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the above embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present application.

[0143] As used in the above embodiments, the term “when…” may be interpreted as “if…” or “after…” or “in response to determining…” or “in response to detecting…”, depending on the context. Similarly, the phrases “upon determining…” or “if (stated condition or event) is detected” may be interpreted as “if determining…” or “in response to determining…” or “upon detecting (stated condition or event)” or “in response to detecting (stated condition or event)”, depending on the context.

[0144] In the above embodiments, all or part of the embodiments can be implemented using software, hardware, firmware, or any combination thereof. When implemented using software, all or part of the embodiments can be implemented in the form of a computer program product. The computer program product includes one or more computer instructions. When the computer program instructions are loaded and executed on a computer, the processes or functions described in the embodiments of the present application are generated in whole or in part. The computer can be a general-purpose computer, a special-purpose computer, a computer network, or other programmable device. The computer instructions can be stored in a computer-readable storage medium or transmitted from one computer-readable storage medium to another. For example, the computer instructions can be transmitted from one website, computer, server, or data center to another website, computer, server, or data center via wired (e.g., coaxial cable, optical fiber, digital subscriber line) or wireless (e.g., infrared, wireless, microwave, etc.) means. The computer-readable storage medium can be any available medium that can be accessed by a computer or a data storage device such as a server or data center that integrates one or more available media. The available medium can be a magnetic medium (e.g., a floppy disk, hard disk, tape), an optical medium (e.g., a DVD), or a semiconductor medium (e.g., a solid-state drive).

[0145] Those skilled in the art will appreciate that all or part of the process steps in the above-described method embodiments can be implemented by a computer program instructing the relevant hardware. The program can be stored in a computer-readable storage medium, and when executed, the program can include the process steps in the above-described method embodiments. The aforementioned storage medium includes various media capable of storing program code, such as ROM or random access memory (RAM), magnetic disks, or optical disks.

Claims

1. A city-level timing strategy optimization method, characterized in that: include: Obtain real-time traffic flow, speed, vehicle type, queue length, and lane occupancy at each intersection collected by multi-source traffic data collection equipment to obtain raw traffic data; Calculating the vehicle delay time at each intersection in different directions using an improved Webster delay formula based on the original traffic data; Classifying the traffic status of each intersection into unimpeded, lightly congested, moderately congested, and severely congested according to the vehicle delay time at each intersection in different directions; Based on the traffic state, a timing strategy optimization model is constructed with minimizing the total delay time as the objective function. The urban road network is abstracted into a graph structure, in which each intersection is a node and each road section is an edge. The constraints of the timing strategy optimization model include a minimum green light time constraint, a maximum green light time constraint, and an intersection saturation constraint. The signal light cycle, phase duration and phase difference of each intersection are used as timing parameters, and the optimal timing solution is obtained by iteratively calculating the solution of the timing strategy optimization model; Sending the optimal timing plan to the traffic light controllers at each intersection, so that all the traffic light controllers operate according to the optimal timing plan; The said constructing a timing strategy optimization model based on the said traffic state with minimizing the total delay time as the objective function specifically includes: calculating the traffic capacity coefficient of each road section based on the traffic state of each intersection; Establishing a road network connectivity constraint matrix, wherein element values of the road network connectivity constraint matrix are determined according to the distance between adjacent intersections, the traffic arrival rate, and the green wave belt speed; The total delay time is expressed as a function of the signal timing parameters of each intersection according to the capacity coefficient and the road network connectivity constraint matrix, wherein the function includes a random delay term and a uniform delay term, wherein the random delay term is determined by the randomness of vehicle arrival and the uniform delay term is determined by the signal timing plan; Establishing mathematical expressions for a minimum green light time constraint, a maximum green light time constraint, and an intersection saturation constraint, wherein the minimum green light time constraint is determined based on pedestrian crossing demand, the maximum green light time constraint is determined based on vehicle queue length, and the intersection saturation constraint is determined based on traffic capacity; The function is used as the objective function, and the minimum green light time constraint, the maximum green light time constraint and the intersection saturation constraint are used as constraint conditions to obtain a timing strategy optimization model.

2. A city-level timing strategy optimization method according to claim 1, characterized in that: The calculation of the vehicle delay time at each intersection in different directions using the improved Webster delay formula based on the original traffic data specifically includes: Preprocessing the raw traffic data to obtain preprocessed traffic data, wherein the preprocessing includes eliminating abnormal data that exceeds a preset range; Dividing the pre-processed traffic data into time windows, and calculating the average vehicle arrival rate and average service time of each intersection within each time window; The average vehicle arrival rate and the average service time are substituted into the improved Webster delay formula to calculate the vehicle delay time of each intersection in different directions.

3. A city-level timing strategy optimization method according to claim 1, characterized in that: The traffic status of each intersection is divided into unimpeded, lightly congested, moderately congested and severely congested according to the vehicle delay time of each intersection in different directions, specifically including: The average delay time of each intersection in different time periods obtained by statistics is used as a benchmark threshold; Calculating the deviation rate between the real-time delay time of each intersection and the benchmark threshold using the benchmark threshold; Constructing a state evaluation index matrix according to the deviation rate, and performing preliminary clustering using the state evaluation index matrix to obtain a clustering result; The traffic status of each intersection is divided into unimpeded, lightly congested, moderately congested and severely congested according to the clustering result.

4. A city-level timing strategy optimization method according to claim 1, characterized in that: After sending the optimal timing plan to the traffic light controllers at each intersection, the method further includes: Constructing a road network period fluctuation curve based on the historical timing data of each intersection, wherein the road network period fluctuation curve reflects the variation pattern of the signal period of each intersection in the road network; Taking the local extreme point of the periodic fluctuation curve in the time dimension as the key time node; Calculating the cycle difference between two adjacent intersections at the key time node; When the period difference is greater than a preset threshold, marking the road section between two adjacent intersections as a potential bottleneck section; The potential bottleneck section is subjected to reverse timing optimization. The reverse timing optimization is to adjust the timing parameters of the downstream intersection inversely based on the upstream intersection, so that the traffic capacity of the potential bottleneck section matches the traffic volume.

5. The method for optimizing city-level timing strategy according to claim 4, characterized in that: The step of constructing a road network periodic fluctuation curve based on the historical timing data of each intersection specifically includes: Using a sliding time window method to process the historical timing data of each intersection in sections; Wavelet transform is used to perform multi-scale decomposition on the data in each time window to extract the main frequency components of periodic changes; The periodic variation trend is reconstructed based on the main frequency components to generate a smooth road network periodic fluctuation curve.

6. A city-level timing strategy optimization method according to claim 4, characterized in that: Calculating the cycle difference between two adjacent intersections at the key time node specifically includes: Establishing a traffic intersection signal timing feature vector, wherein the traffic intersection signal timing feature vector includes a traffic intersection signal timing feature vector cycle length, a green-to-signal ratio, a phase difference, and a flow direction distribution; Calculating the feature vector distance between the timing feature vectors of two adjacent intersections using a multi-dimensional feature distance calculation method based on improved cosine similarity; Calculate the road section saturation weight coefficient based on the road section capacity and actual demand; The eigenvector distance is multiplied by the saturation weight coefficient to obtain the period difference.

7. A city-level timing strategy optimization system, characterized in that: The system comprises: 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 system to execute a city-level timing strategy optimization method as described in any one of claims 1-6.

8. A computer-readable storage medium comprising instructions, characterized in that: When the instruction is executed on the system, the system is caused to execute the city-level timing strategy optimization method according to any one of claims 1 to 6.

9. A computer program product, characterized in that When the computer program product is run on a system, the system is enabled to execute a city-level timing strategy optimization method according to any one of claims 1 to 6.

Images