Macro-micro traffic regulation method and device based on optimization solver, electronic equipment and storage medium

Abstract

The application provides a macro-micro traffic regulation method and device based on an optimization solver, electronic equipment and a storage medium, and relates to the technical field of traffic regulation, wherein the method comprises: acquiring multi-source traffic data of a target area; calling a macro model to determine an origin-destination demand matrix according to the multi-source traffic data; inputting the origin-destination demand matrix into a meso model to generate traffic flow parameters; taking the traffic flow parameters as input conditions of a micro model, constructing a mixed integer programming model based on the traffic flow parameters, solving the mixed integer programming model based on a preset solver strategy, and obtaining a traffic control scheme; updating the road network operation state based on the traffic control scheme, and determining the current road network state as an optimal road network state when the deviation rate of the updated road network state and real-time collected traffic monitoring data does not exceed a preset threshold. The application can select an adaptive solver for solving according to the actual problem size, and can further adapt to the traffic regulation requirements in different scenarios.

Description

Technical Field

[0001] This application relates to the field of traffic control technology, and in particular to a macro-micro traffic control method, device, electronic device and storage medium based on an optimization solver. Background Technology

[0002] With the acceleration of urbanization and the continuous growth of motor vehicle ownership, problems such as traffic congestion, environmental pollution and declining travel efficiency have become increasingly prominent. As an important means of alleviating these problems, traffic control, in particular, is directly related to the efficiency of urban operation and the travel experience of residents through its level of intelligence and control precision.

[0003] Existing traffic control methods typically use fixed solvers to obtain corresponding control schemes, making it impossible to select an appropriate solver based on the actual problem scale, and thus difficult to adapt to traffic control needs in different scenarios. Summary of the Invention

[0004] This application provides a macro-micro traffic control method, device, electronic device, and storage medium based on an optimization solver, which solves the technical problem that the existing technology usually uses a fixed solver to obtain the corresponding control scheme, and cannot select an appropriate solver according to the actual problem scale, making it difficult to adapt to the traffic control needs in different scenarios.

[0005] This application provides a macro-micro traffic control method based on an optimization solver, including: Acquire multi-source traffic data for the target area; wherein the multi-source traffic data includes at least one of historical travel data, population employment attributes, and origin-destination data; The traffic control scheme generation steps are as follows: A macro-level model is invoked to determine the origin-destination demand matrix based on the multi-source traffic data; the origin-destination demand matrix is ​​input into a meso-level model, and a dynamic traffic assignment algorithm is used to generate traffic flow parameters; a mixed-integer programming model is constructed based on the traffic flow parameters, and the mixed-integer programming model is solved based on a preset solver strategy to obtain the traffic control scheme; wherein, the traffic control scheme includes a signal control scheme and a vehicle scheduling scheme; the preset solver strategy is determined based on the historical performance of each solver in solving the mixed-integer programming model; The optimal road network state determination steps are as follows: based on the traffic control scheme, the road network operation state is updated, the deviation rate between the updated road network state and the real-time collected traffic monitoring data is determined, and if the deviation rate does not exceed a preset threshold, the current road network state is determined as the optimal road network state.

[0006] According to the macro-micro traffic control method based on an optimization solver provided in this application, the step of determining the origin-destination demand matrix based on the multi-source traffic data includes: The target area is divided into traffic zones, resulting in multiple traffic zones. Establish the spatial mapping relationship between the traffic zones and road network nodes; Based on the historical travel data, population employment attributes, and origin-destination data of the aforementioned traffic zones, the origin-destination demand matrix of each traffic zone is predicted. Based on the spatial mapping relationship, the origin-destination demand matrix of each traffic zone is converted into the origin-destination demand matrix between the corresponding road network nodes.

[0007] According to the macro-micro traffic control method based on an optimization solver provided in this application, the step of constructing a mixed integer programming model based on the traffic flow parameters, solving the mixed integer programming model based on a preset solver strategy, and obtaining a traffic control scheme includes: The signal cycle length and the green light duration for each phase are used as the first decision variables; Based on the traffic flow parameters, a first objective function is constructed with the goal of minimizing the average vehicle delay at the intersection. The constraints of the first objective function include periodic constraints, green light duration constraints, total number of green lights constraints, and phase sequence constraints. Construct a first mixed-integer programming model based on the first decision variable and the first objective function; The first mixed integer programming model is solved using a preset solver strategy to obtain a signal control scheme.

[0008] According to the macro-micro traffic control method based on an optimization solver provided in this application, the step of constructing a mixed integer programming model based on the traffic flow parameters and solving the mixed integer programming model based on a preset solver strategy to obtain a traffic control scheme further includes: The passenger set, station set, and vehicle set are used as the second decision variables; A second objective function is constructed based on the fixed costs of the vehicle and the variable costs of the vehicle serving passengers at the station; A second mixed-integer programming model is constructed based on the second decision variable and the second objective function, wherein the constraints of the second mixed-integer programming model include capacity constraints, time window constraints, and service continuity constraints. The mixed integer programming model is solved using a preset solver strategy to generate a vehicle scheduling scheme; wherein the vehicle scheduling scheme includes a vehicle departure plan, a stop sequence, and a passenger allocation scheme.

[0009] According to the macro-micro traffic control method based on optimization solver provided in this application, the determination of the preset solver strategy includes: Determine the task type of the mixed integer programming model, and query the historical performance of each solver based on the task type; The overall score for each solver is determined based on its historical performance. Select the preset quantity solver with the highest comprehensive score.

[0010] According to the macro-micro traffic control method based on an optimization solver provided in this application, the updated road network state includes the average vehicle speed of road segments, and determining the deviation rate between the updated road network state and the real-time collected traffic monitoring data includes: Real-time data collection of average vehicle speed and the number of monitored road segments; The deviation rate is determined based on the average vehicle speed, the number of monitored road segments, and the average vehicle speed of the road segments.

[0011] According to the macro-micro traffic control method based on an optimization solver provided in this application, the step of updating the road network operating state based on the traffic control scheme and determining the deviation rate between the updated road network state and the real-time collected traffic monitoring data further includes: If the deviation rate exceeds a preset threshold, the model parameters of the mesoscopic model are adjusted, and the traffic control scheme generation step and the optimal road network state determination step are repeated.

[0012] This application also provides a macro-micro traffic control device based on an optimization solver, comprising: A multi-source traffic data acquisition module is used to acquire multi-source traffic data of a target area; wherein, the multi-source traffic data includes at least one of historical travel data, population employment attributes, and origin-destination data; A traffic control scheme generation module is used to call a macro-level model to determine the origin-destination demand matrix based on the multi-source traffic data; input the origin-destination demand matrix into a meso-level model, and generate traffic flow parameters using a dynamic traffic assignment algorithm; construct a mixed-integer programming model based on the traffic flow parameters, and solve the mixed-integer programming model based on a preset solver strategy to obtain a traffic control scheme; wherein, the traffic control scheme includes a signal control scheme and a vehicle scheduling scheme; the preset solver strategy is determined based on the historical performance of each solver in solving the mixed-integer programming model; The optimal road network state determination module is used to update the road network operation state based on the traffic control scheme, determine the deviation rate between the updated road network state and the real-time collected traffic monitoring data, and determine the current road network state as the optimal road network state if the deviation rate does not exceed a preset threshold.

[0013] This application also provides an electronic device, including a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein the processor executes the computer program to implement the macro-micro traffic control method based on the optimization solver as described above.

[0014] This application also provides a non-transitory computer-readable storage medium having a computer program stored thereon, which, when executed by a processor, implements the macro-micro traffic control method based on an optimization solver as described above.

[0015] This application also provides a computer program product, including a computer program that, when executed by a processor, implements the macro-micro traffic control method based on an optimization solver as described above.

[0016] This application constructs a mixed integer programming model based on the generated traffic flow parameters, and determines the current solver strategy based on the historical performance of each solver in solving the integer programming model. Thus, the current mixed integer programming model is solved according to the current solver strategy. It can select an appropriate solver to solve the problem according to the actual problem size, thereby adapting to the traffic control needs in different scenarios.

[0017] This application automatically inputs the origin-destination demand matrix determined by the macro model into the meso model. The traffic flow parameters generated by the meso model are automatically used as input conditions for the micro model, thereby generating a traffic control scheme. It constructs an automatic serial call link from macro model to meso model to micro model, eliminating the cumbersome process of manual export, format conversion, and re-import. It realizes the automatic transfer of data between the three-layer model, ensuring that the generation of traffic control schemes is based on accurate and continuous data flow, thereby significantly improving the effectiveness of traffic control.

[0018] Furthermore, when the deviation rate exceeds a preset threshold, this application adjusts the model parameters of the meso-level model and repeatedly executes the traffic control scheme generation step and the optimal road network state determination step. This can improve the accuracy of meso-level allocation and micro-level control from top to bottom, forming a complete three-level linkage closed loop of micro-level feedback → meso-level adjustment → macro-level correction, thereby effectively improving the accuracy of traffic regulation. Attached Figure Description

[0019] To more clearly illustrate the technical solutions in this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0020] Figure 1 This is a flowchart illustrating the macro-micro traffic control method based on an optimization solver provided in this application.

[0021] Figure 2 This is a schematic diagram of the macro-meta-micro model computation framework provided in this application.

[0022] Figure 3 This is a schematic diagram of the macro-micro traffic control device based on an optimization solver provided in this application.

[0023] Figure 4 This is a schematic diagram of the structure of the electronic device provided in this application. Detailed Implementation

[0024] To make the objectives, technical solutions, and advantages of this application clearer, the technical solutions of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0025] Figure 1 This is a flowchart illustrating the macro-micro traffic control method based on an optimization solver provided in this application, as shown below. Figure 1 As shown, the method includes the following: S1. Obtain multi-source traffic data for the target area; wherein, the multi-source traffic data includes at least one of historical travel data, population employment attributes, and origin-destination data; This application's embodiments can be applied to traffic control platforms that access multi-source traffic data through a single data interface, including population employment attributes, origin-destination data generated by mobile signaling, traffic detector flow data, signal timing data, and some video trajectory data. After access, various data types can be preprocessed to form a unified data format that can be directly used. Preprocessing may include format conversion, field standardization, and coordinate unification.

[0026] S2. Traffic Control Scheme Generation Steps: The macro-level model is invoked to determine the origin-destination demand matrix based on the multi-source traffic data; the origin-destination demand matrix is ​​input into the meso-level model, and a dynamic traffic assignment algorithm is used to generate traffic flow parameters; a mixed-integer programming model is constructed based on the traffic flow parameters, and the mixed-integer programming model is solved based on a preset solver strategy to obtain the traffic control scheme; wherein, the traffic control scheme includes a signal control scheme and a vehicle scheduling scheme; the preset solver strategy is determined based on the historical performance of each solver in solving the mixed-integer programming model; In this embodiment of the application, traffic flow parameters include at least one of the following: passenger set, station set, vehicle set signal cycle length, and green light duration for each phase.

[0027] S3. Optimal road network state determination steps: Update the road network operation state based on the traffic control scheme, determine the deviation rate between the updated road network state and the real-time collected traffic monitoring data, and determine the current road network state as the optimal road network state if the deviation rate does not exceed a preset threshold.

[0028] In this embodiment, the origin-destination demand matrix determined by the macro model is automatically input into the meso model. The traffic flow parameters generated by the meso model are automatically used as input conditions for the micro model, thereby generating a traffic control scheme. This constructs an automatic serial calling link from macro model to meso model to micro model, eliminating the cumbersome process of manual export, format conversion, and re-import. It realizes the automated transfer of data between the three-layer model, ensuring that the generation of traffic control schemes is based on accurate and continuous data flow, thereby significantly improving the effectiveness of traffic regulation.

[0029] It should be noted that in the field of traffic control technology, macroscopic models, mesoscopic models, and microscopic models are three classic types of modeling for describing the operational state of traffic flow. Macroscopic models describe the collective behavior of traffic flow from a holistic perspective, treating it as a compressible continuous fluid. They focus on the overall parameters of traffic flow (flow rate, density, speed) and their spatiotemporal changes, without paying attention to the motion details of individual vehicles. Mesoscopic models strike a balance between macroscopic and microscopic approaches, modeling vehicles in groups (e.g., vehicle groups, vehicle packages), focusing on the behavior of the vehicle group but ignoring the specific driving operations of individual vehicles. Microscopic models describe the motion behavior of each vehicle from an individual perspective, focusing on the interaction details between vehicles and between vehicles and road infrastructure, and can simulate specific driving operations such as lane changes, following, overtaking, and traffic light responses.

[0030] Please see Figure 2 In one embodiment, a schematic diagram of a macro-meta-micro model computation framework is provided. For example... Figure 2 As shown, the macro model performs OD (Origin-Destination) demand forecasting and transmits the forecast results to the meso model. The meso model performs traffic flow allocation and transmits the allocation results to the micro model. The micro model then performs path coordination optimization and traffic light control to achieve traffic regulation.

[0031] This application embodiment constructs a mixed integer programming model based on the generated traffic flow parameters, and determines the current solver strategy based on the historical performance of each solver in solving the integer programming model. Thus, the current mixed integer programming model is solved according to the current solver strategy. It can select an appropriate solver to solve the problem according to the actual problem size, thereby adapting to the traffic control needs in different scenarios.

[0032] In one embodiment, step S2, determining the origin-destination demand matrix based on the multi-source traffic data, includes: S211. Divide the target area into traffic zones to obtain multiple traffic zones; In this embodiment, the target area can be divided into several non-overlapping traffic zones based on its road topology, administrative boundaries, land use characteristics, and population distribution. These traffic zones are the basic spatial units for macro-level traffic demand forecasting, and their granularity is determined according to the required research precision. For example, larger-scale zones can be used for city-level planning, while smaller-scale zones are used for district-level fine-tuning. After division, each traffic zone is assigned a unique identifier, and its geometric boundaries, centroid coordinates, and land use attributes (such as residential, commercial, or industrial areas) are recorded.

[0033] S212. Establish the spatial mapping relationship between the traffic zones and road network nodes; In this embodiment, the centroid coordinates of each traffic zone can be mapped to the nearest road network node (i.e., intersection or road segment endpoint) in the road network, forming a "zone-node" correspondence table, which is the spatial mapping relationship. This spatial mapping relationship is a key bridge connecting macro-level zone-level demand and meso-level road network-level calculation. Specifically, it is implemented as follows: based on road network topology data, the Euclidean distance or network distance from the centroid of each zone to surrounding road network nodes is calculated, and the nearest node is selected as the attached node for that zone. For special cases (such as a zone spanning multiple nodes), area-weighted or multi-node mapping strategies can be used. After mapping, an association index between zone IDs and node IDs is established to ensure that the subsequent OD matrix can be accurately converted into travel demand between node pairs.

[0034] S213. Based on the historical travel data, population employment attributes, and origin-destination data of the traffic zones, predict the origin-destination demand matrix of each traffic zone; In this embodiment of the application, based on historical travel data, population employment attributes, and origin-destination data of each traffic zone, the travel generation model and travel distribution model in the macroscopic fourth-order method are used to predict the travel exchange volume between zones and determine the origin-destination demand matrix, specifically: Using cross-classification or regression analysis, the trip generation (trips originating from home) and trip attraction (trips ending at home) of each community are calculated based on the population and employment attributes of each community, with the unit being "person-trips / day".

[0035] Constrained by the generation and attraction of each cell, iterative equilibrium calculations are performed using a gravity model (considering distance between cells, travel time, or generalized impedance) or the Frata method to obtain the travel exchange volume between cell pairs, ultimately forming a two-dimensional origin-destination demand matrix (OD matrix). The rows of this matrix represent the origin cells, and the columns represent the destination cells.

[0036] S214. Based on the spatial mapping relationship, the origin-destination demand matrix of each traffic zone is converted into the origin-destination demand matrix between the corresponding road network nodes.

[0037] In this embodiment of the application, converting the cell-level OD matrix into a node-level OD matrix based on spatial mapping relationships may include: For each pair of cells in the OD matrix, its travel volume is mapped to the travel demand between the corresponding attached node pairs; if multiple cells are mapped to the same node, the total generation and attraction of that node are summed. The resulting OD demand matrix between nodes preserves the spatiotemporal distribution characteristics of the original cell OD, while also adapting to the input requirements of the mesoscopic model with road network nodes as origin and destination points.

[0038] This application's embodiments comprehensively predict the inter-regional origin-destination demand matrix by integrating multi-source information such as historical travel data, population employment attributes, and original origin-destination data. This multi-source data fusion mechanism can more comprehensively reflect the characteristics of regional travel generation and attraction. Compared with a single data source, it effectively improves the accuracy of travel demand prediction and provides reliable data support for subsequent meso-level dynamic traffic allocation.

[0039] In one embodiment, step S2, constructing a mixed-integer programming model based on the traffic flow parameters, and solving the mixed-integer programming model based on a preset solver strategy to obtain a traffic control scheme, includes: S221. The signal cycle length and the green light duration of each phase are used as the first decision variables. In this embodiment of the application, an intersection signal timing optimization model can be constructed based on traffic flow parameters, and a traffic control scheme can be generated based on the intersection timing optimization model. The input data of the intersection timing optimization model can be the traffic flow and saturation flow of each approach lane of the target intersection, as well as the current phase scheme (number of phases P, combination of approach lanes included in each phase).

[0040] The first decision variables include the signal period length C (seconds) and the green light duration g for each phase. p (p=1,...,P). Green light ratio λ for each phase. p =g p / C is a derived variable.

[0041] S222. Based on the traffic flow parameters, construct a first objective function with the goal of minimizing the average vehicle delay at the intersection, wherein the constraints of the first objective function include periodic constraints, green light duration constraints, total number of green lights constraints, and phase sequence constraints. In this embodiment, the expression of the first objective function is as follows:

[0042] Where d is the average vehicle delay at the intersection, and y p =q p / s p Let λ be the flow saturation at phase p, C be the signal period, and λ be the flow rate saturation. p For green credit ratio.

[0043] The constraints include ① periodic constraints: C min ≤C≤C max ② Green light duration constraint: g p ≥g min (Minimum green light time); ③ Total green light limit: Σ p g p =CL ( L (Total lost time); ④ Phase sequence constraint: Keep the original phase sequence unchanged.

[0044] S223. Construct a first mixed-integer programming model based on the first decision variable and the first objective function; S224. Solve the first mixed integer programming model using a preset solver strategy to obtain a signal control scheme.

[0045] In this embodiment, a preset solver strategy can be used to solve the first mixed-integer programming model constructed by the first objective function, outputting the optimal value of the green light duration for each phase, and generating a signal control scheme based on the optimal value of the green light duration for each phase, including the signal period C and the green light duration g for each phase. p The signal control scheme, which includes phase switching sequence and yellow light duration (fixed parameters), can be directly sent to the simulation platform or actual signal control equipment for execution.

[0046] This application embodiment uses the traffic flow parameters output from the mesoscopic model as the input to the signal timing optimization model, ensuring that the generation of the signal control scheme is strictly based on the real-time road network operating status, rather than relying on human experience or fixed timing. The first objective function focuses on minimizing the average vehicle delay at the intersection, and is optimized by combining multiple constraints such as cycle time and green light duration. This results in the output of timing parameters that accurately match the current traffic demand, significantly improving intersection efficiency and reducing vehicle delays. Furthermore, the solution results of the signal timing optimization model in this application embodiment can be directly transmitted back to the platform database to update the road network status, providing a basis for subsequent deviation calibration and parameter correction, avoiding the problem of signal timing being out of sync with upper-level requirements and requiring manual intervention for adjustment, and realizing the integrated linkage of macro, meso, and micro three-level control.

[0047] In one embodiment, step S2, which involves constructing a mixed-integer programming model based on the traffic flow parameters and solving the mixed-integer programming model based on a preset solver strategy to obtain a traffic control scheme, further includes: S231. Use the passenger set, station set, and vehicle set as the second decision variables; In this embodiment of the application, the expression for the second decision variable is as follows: in Gather passengers. For site collection, For vehicle assembly, Indicates vehicle On the site Serving passengers .

[0048] S232. Construct a second objective function based on the fixed costs of the vehicle and the variable costs of the vehicle serving passengers at the station; In this embodiment of the application, the expression of the second objective function is as follows:

[0049] in, For vehicles Fixed costs, For vehicles On the site Serving passengers Variable costs.

[0050] S233. Construct a second mixed-integer programming model based on the second decision variable and the second objective function, wherein the constraints of the second mixed-integer programming model include capacity constraints, time window constraints, and service continuity constraints. In this embodiment of the application, the expression for the constraint condition is as follows: Capacity constraints:

[0051] in For passengers The number of people, For vehicles The maximum capacity.

[0052] Time window constraints: in For vehicles Arrival Station Time, This refers to the available service time window.

[0053] Service continuity constraints: in, For vehicles Arrival Station Time, For service hours, Travel time between stations For vehicle k to reach the next station The time.

[0054] S234. Solve the second mixed integer programming model using a preset solver strategy to generate a vehicle scheduling scheme; wherein, the vehicle scheduling scheme includes a vehicle departure plan, a stop sequence, and a passenger allocation scheme.

[0055] This application embodiment constructs a second mixed integer programming model by using the passenger set, station set, and vehicle set as the second decision variables. This transforms the vehicle scheduling problem into a mathematical optimization problem. The model comprehensively considers capacity constraints, time window constraints, and service continuity constraints. It can simultaneously optimize the vehicle departure plan, stop sequence, and passenger allocation scheme, and automatically generate globally optimal scheduling instructions.

[0056] Furthermore, this application embodiment employs a preset solver strategy to solve mixed integer programming models. It can automatically match the optimal solver based on characteristics such as problem size and constraint complexity, and supports parallel solving by multiple solvers and result comparison. This mechanism solves the problems of fixed solvers being unable to adapt to problems of different sizes, having low solving efficiency, or being prone to failure in traditional methods. It ensures that computation time is shortened as much as possible while maintaining solution accuracy, meeting the rapid response requirements in real-time scheduling scenarios.

[0057] In one embodiment, determining the preset solver strategy includes: Determine the task type of the mixed integer programming model, and query the historical performance of each solver based on the task type; In this embodiment of the application, the historical performance of various candidate solvers on similar tasks can be queried from the solver performance database, including indicators such as average solution time, solution success rate, and stability of objective function value, as the basis data for subsequent scoring.

[0058] The overall score for each solver is determined based on its historical performance. In this embodiment of the application, the expression for the comprehensive score is as follows: in, For comprehensive scoring, Score points for speed. For stability score, To score for accuracy, For the weighting coefficients, satisfying .

[0059] Speed ​​Score:

[0060] in For solver The average solution time for similar problems, The maximum and minimum times are among all solvers.

[0061] Stability score:

[0062] Where M represents the total number of solutions. For the first The objective function value is then solved. This is the average value. The allowable deviation threshold, This is an indicator function.

[0063] Select the preset quantity solver with the highest comprehensive score.

[0064] In this embodiment, the top three solvers with the highest comprehensive scores can be selected to solve the relevant objective function. For example, for a vehicle scheduling task, based on a preset solver strategy, three mainstream solvers are simultaneously invoked. The objective function value, solution time, result deviation rate, and stability of multiple solutions for each solver are recorded. The first solver has an objective function value of a certain value, a solution time of approximately 225 seconds, a deviation rate of approximately 1.2%, and a result stability variance of approximately 0.03. The second solver has an objective function value of a certain value, a solution time of approximately 198 seconds, a deviation rate of zero, and a result stability variance of approximately 0.01. The third solver has an objective function value of a certain value, a solution time of approximately 210 seconds, a deviation rate of approximately 0.3%, and a result stability variance of approximately 0.01.

[0065] This application's embodiments can construct a multi-solver selection engine and a distributed algorithm collaborative computing framework to achieve dynamic solver switching, parallel computation of multiple algorithms, and result comparison, including: An automatic problem feature extraction mechanism is designed. By identifying key features of the optimization model, a multi-objective scoring and matching algorithm is established. Based on the problem size, the historical performance of the solver at that size is queried, and the computation speed is matched with a score. The stability of the solver is evaluated based on its historical success rate; solvers with high success rates receive higher stability scores, and the top three solvers with the highest scores are recommended to the user as the preferred options. An intelligent task decomposition strategy is designed to break down complex tasks into parallelizable subtasks based on task type. For regional signal optimization tasks, intersections within the region are grouped by spatial location, and each group of intersections is assigned as an independent subtask to different slave nodes. For example, for travel demand prediction tasks, the same problem is assigned to multiple algorithms for parallel computation to obtain diverse results.

[0066] In this embodiment, multiple types of callable algorithm models and solver resources can be pre-integrated, including mathematical programming algorithms for combinatorial optimization problems, heuristic algorithms for large-scale search problems, data-driven models for prediction problems, and corresponding different solution execution environments. All algorithms and solvers are encapsulated into schedulable computing units within the traffic scheduling platform through a unified interface, and a mapping relationship is established between algorithm type, applicable problem size, input data structure, and execution characteristics.

[0067] Users can select a computational scheme from the configured algorithm and solver library that matches the current problem scenario based on the specific computational task. After selection, the traffic scheduling platform automatically completes data adaptation and task scheduling, inputting structured data into the corresponding method module, and the matching method executes the core computation process. The intermediate states and final results generated during execution are transmitted back to the platform's data layer in real time through a unified data interface. After the platform performs structured processing on the returned results, it transmits them to the visualization module for display, and they can also be used as input data for subsequent model building and task scheduling.

[0068] Furthermore, the traffic dispatch platform provides a front-end interface for users to operate. When operating on the front-end interface, users can obtain operational information at different scales through querying, filtering, or regional selection. Data requests generated by user interactions are transmitted to the platform's computing module through an interface. The computing module dynamically calls relevant data or re-executes the calculation process according to the request and feeds back the results to the front-end display module in real time.

[0069] In one embodiment, the updated road network status includes the average vehicle speed of road segments, and determining the deviation rate between the updated road network status and the real-time collected traffic monitoring data includes: Real-time data collection of average vehicle speed and the number of monitored road segments; The deviation rate is determined based on the average vehicle speed, the number of monitored road segments, and the average vehicle speed of the road segments.

[0070] In this embodiment of the application, the expression for the deviation rate is as follows:

[0071] in, The deviation rate, For the simulated road section average speed, The measured average speed, To monitor the number of road sections.

[0072] This application embodiment collects and monitors the average vehicle speed of the road segment in real time, compares it with the average vehicle speed of the road segment output by simulation, and calculates a quantitative deviation rate index. This provides a unified error evaluation scale for models at the macro, meso, and micro levels, and provides a clear trigger basis for subsequent closed-loop calibration, which is conducive to improving the accuracy of vehicle scheduling.

[0073] In one embodiment, step S3, updating the road network operating status based on the traffic control scheme and determining the deviation rate between the updated road network status and the real-time collected traffic monitoring data, further includes: If the deviation rate exceeds a preset threshold, the model parameters of the mesoscopic model are adjusted, and the traffic control scheme generation step and the optimal road network state determination step are repeated.

[0074] In this embodiment of the application, when the deviation rate exceeds a preset threshold, the model parameters of the mesoscopic model can be adjusted, including adjusting the model parameters of the road segment impedance function in the mesoscopic model. The expression is as follows:

[0075] in, The actual impedance of road segment a. Let be the free-flow time of road segment a. This refers to the actual traffic flow on road segment a. It refers to the traffic capacity of road segment a. It is the impedance coefficient. It is the impedance index.

[0076] In this embodiment, the parameters in the road segment impedance function can be adjusted. This allows the simulated traffic flow to match the measured traffic flow. In this embodiment, the mesoscopic model is fine-tuned, and based on the fine-tuned model, the traffic control scheme generation step and the optimal road network state determination step are repeatedly executed. The deviation rate is recalculated, and if the deviation rate does not exceed a preset threshold, the optimal road network state and the optimal traffic control scheme are output.

[0077] In this embodiment, when the deviation rate originates from a specific region and is related to the recent implementation of traffic policies, the policy impact coefficient in the macro model can also be corrected. For example, if the actual traffic flow decreases far more than the model predicts after congestion pricing is implemented in a certain region, the platform will dynamically enhance the sensitivity coefficient of the pricing factor in the road resistance function of that region, making the model's evaluation of such policies more accurate in the future.

[0078] When the deviation is identified as a regional distribution deviation, the impedance coefficients in the gravity model or distribution model will be adjusted to rebalance the attraction weights between traffic zones, thereby indirectly correcting the spatial distribution structure of the OD matrix. When the deviation is identified as a global total deviation, the input origin-destination demand matrix will be directly scaled proportionally or corrected using a more complex matrix estimation method.

[0079] In this embodiment, when the deviation rate exceeds a preset threshold, the model parameters of the meso-level model are adjusted, and the traffic control scheme generation step and the optimal road network state determination step are repeatedly executed. This can improve the accuracy of meso-level allocation and micro-level control from top to bottom, forming a complete three-level linkage closed loop of micro-level feedback → meso-level adjustment → macro-level correction, thereby effectively improving the accuracy of traffic regulation.

[0080] In one embodiment, the real-time calculation results can also be visualized through a visual interface.

[0081] In this embodiment, the visualization interface includes a real-time dynamic update mechanism, platform panel design, computation progress visualization, and interactive analysis functions. The real-time dynamic update mechanism establishes a complete data flow channel from the computing nodes to the front-end visualization client. The computing nodes generate real-time computation results and encapsulate them into a standard data format, which is then buffered and distributed through a message queue system. A network socket server establishes a persistent connection with the browser client, pushing incrementally updated data to the front-end visualization module in real time.

[0082] Platform panel design: It adopts a left-center-right column layout. The main area displays a digital twin 3D map, occupying a large display space and providing an immersive spatial display experience. The left and right auxiliary areas display key indicator monitoring panels, showcasing real-time calculation results in various formats such as cards, charts, and tables.

[0083] Visualization of progress calculations: Real-time indicator cards display core indicator values, including key performance metrics such as average delay time, queue length, and throughput efficiency. Time series graphs are drawn using a professional charting library, showing the dynamic trends of key indicators over time. The horizontal axis represents time, and the vertical axis represents indicator values. New data points are added to the curve in real time, forming a continuous trajectory of change. Multiple curves can be overlaid for easy comparison and analysis of different indicators or different solutions.

[0084] Interactive analysis functionality: Users can select their desired parameters, scenarios, algorithms, etc., via a panel, and the system responds promptly to user actions and displays the results on the page. It supports algorithm result comparison; the algorithm performance comparison table displays comparative data for various algorithms in terms of accuracy, computation time, resource consumption, etc., in tabular form. It also supports sorting by different dimensions to quickly identify the optimal algorithm.

[0085] In one embodiment, macro-micro traffic control based on an optimization solver is applicable to a traffic dispatching platform. This platform includes a data interface layer, a model building layer, a solver calling layer, and a result feedback layer. The data interface layer uses a RESTful API as the data interaction channel, supporting both JSON and CSV data formats to ensure data transmission compatibility and universality. It collects core passenger reservation data, including passenger origin and destination latitude and longitude, station travel time, and passenger numbers. It also queries operational vehicle status data, including current vehicle location, remaining passenger capacity, and information on passengers already served. The model building layer generates a binary decision variable matrix based on the combination relationship between stations and vehicles, and generates key constraints such as capacity constraints and time window constraints according to the flexible public transport operation rules. The completed mixed integer programming model is output as a linear integer programming format file to adapt to the input requirements of the optimization solver. Solver Calling Layer: This layer unifies and encapsulates the calling methods of mainstream solvers (including CMIP, CPLEX, and Gurobi) into a standardized calling interface. This reduces solver switching costs and optimizes parameter configurations such as solution time limits, accuracy requirements, and the number of parallel computing threads. It supports active interruption during the solution process, allowing resumption from the breakpoint, thus improving the flexibility and stability of the solution process. Result Feedback Layer: This layer parses and transforms the optimization results output by the solver, generating scheduling instructions that can be directly executed by the simulation platform and actual operating systems, and accurately issuing these instructions. Based on the parsed results, it generates multi-dimensional executable scheduling schemes, including vehicle departure plans, stop sequences, and passenger allocation schemes.

[0086] Implementing the embodiments of this application has the following beneficial effects: This application embodiment constructs a mixed integer programming model based on the generated traffic flow parameters, and determines the current solver strategy based on the historical performance of each solver in solving the integer programming model. Thus, the current mixed integer programming model is solved according to the current solver strategy. It can select an appropriate solver to solve the problem according to the actual problem size, thereby adapting to the traffic control needs in different scenarios.

[0087] In this embodiment, the origin-destination demand matrix determined by the macro model is automatically input into the meso model. The traffic flow parameters generated by the meso model are automatically used as input conditions for the micro model, thereby generating a traffic control scheme. This constructs an automatic serial calling link from macro model to meso model to micro model, eliminating the cumbersome process of manual export, format conversion, and re-import. It realizes the automated transfer of data between the three-layer model, ensuring that the generation of traffic control schemes is based on accurate and continuous data flow, thereby significantly improving the effectiveness of traffic regulation.

[0088] Furthermore, in this embodiment of the application, when the deviation rate exceeds a preset threshold, the model parameters of the meso-level model are adjusted, and the traffic control scheme generation step and the optimal road network state determination step are repeatedly executed. This can improve the accuracy of meso-level allocation and micro-level control from top to bottom, forming a complete three-level linkage closed loop of micro-level feedback → meso-level adjustment → macro-level correction, thereby effectively improving the accuracy of traffic regulation.

[0089] The macro-micro traffic control device based on an optimization solver provided in this application is described below. The macro-micro traffic control device based on an optimization solver described below can be referred to in correspondence with the macro-micro traffic control method based on an optimization solver described above.

[0090] Please see Figure 3 This is a schematic diagram of a macro-micro traffic control device based on an optimization solver, provided in an embodiment of this application. The macro-micro traffic control device based on an optimization solver includes: The multi-source traffic data acquisition module 310 is used to acquire multi-source traffic data of a target area; wherein, the multi-source traffic data includes at least one of historical travel data, population employment attributes, and origin-destination data; The traffic control scheme generation module 320 is used to call the macro model, determine the origin-destination demand matrix based on the multi-source traffic data, input the origin-destination demand matrix into the meso model, generate traffic flow parameters using a dynamic traffic assignment algorithm, construct a mixed-integer programming model based on the traffic flow parameters, and solve the mixed-integer programming model based on a preset solver strategy to obtain a traffic control scheme. The traffic control scheme includes a signal control scheme and a vehicle scheduling scheme. The preset solver strategy is determined based on the historical performance of each solver in solving the mixed-integer programming model. The optimal road network state determination module 330 is used to update the road network operation state based on the traffic control scheme, determine the deviation rate between the updated road network state and the real-time collected traffic monitoring data, and determine the current road network state as the optimal road network state if the deviation rate does not exceed a preset threshold.

[0091] In one embodiment, determining the origin-destination demand matrix based on the multi-source traffic data includes: The target area is divided into traffic zones, resulting in multiple traffic zones. Establish the spatial mapping relationship between the traffic zones and road network nodes; Based on the historical travel data, population employment attributes, and origin-destination data of the aforementioned traffic zones, the origin-destination demand matrix of each traffic zone is predicted. Based on the spatial mapping relationship, the origin-destination demand matrix of each traffic zone is converted into the origin-destination demand matrix between the corresponding road network nodes.

[0092] In one embodiment, the step of constructing a mixed-integer programming model based on the traffic flow parameters, and solving the mixed-integer programming model based on a preset solver strategy to obtain a traffic control scheme includes: The signal cycle length and the green light duration for each phase are used as the first decision variables; Based on the traffic flow parameters, a first objective function is constructed with the goal of minimizing the average vehicle delay at the intersection. The constraints of the first objective function include periodic constraints, green light duration constraints, total number of green lights constraints, and phase sequence constraints. Construct a first mixed-integer programming model based on the first decision variable and the first objective function; The first mixed integer programming model is solved using a preset solver strategy to obtain a signal control scheme.

[0093] In one embodiment, the step of constructing a mixed-integer programming model based on the traffic flow parameters, solving the mixed-integer programming model based on a preset solver strategy, and obtaining a traffic control scheme further includes: The passenger set, station set, and vehicle set are used as the second decision variables; A second objective function is constructed based on the fixed costs of the vehicle and the variable costs of the vehicle serving passengers at the station; A second mixed-integer programming model is constructed based on the second decision variable and the second objective function, wherein the constraints of the second mixed-integer programming model include capacity constraints, time window constraints, and service continuity constraints. The mixed integer programming model is solved using a preset solver strategy to generate a vehicle scheduling scheme; wherein the vehicle scheduling scheme includes a vehicle departure plan, a stop sequence, and a passenger allocation scheme.

[0094] In one embodiment, determining the preset solver strategy includes: Determine the task type of the mixed integer programming model, and query the historical performance of each solver based on the task type; The overall score for each solver is determined based on its historical performance. Select the preset quantity solver with the highest comprehensive score.

[0095] In one embodiment, the updated road network status includes the average vehicle speed of road segments, and determining the deviation rate between the updated road network status and the real-time collected traffic monitoring data includes: Real-time data collection of average vehicle speed and the number of monitored road segments; The deviation rate is determined based on the average vehicle speed, the number of monitored road segments, and the average vehicle speed of the road segments.

[0096] In one embodiment, the step of updating the road network operating status based on the traffic control scheme and determining the deviation rate between the updated road network status and the real-time collected traffic monitoring data further includes: If the deviation rate exceeds a preset threshold, the model parameters of the mesoscopic model are adjusted, and the steps of the traffic control scheme generation module and the optimal road network state determination module are repeated.

[0097] Figure 4 An example is a schematic diagram of the physical structure of an electronic device, such as... Figure 4 As shown, the electronic device may include: a processor 410, a communication interface 420, a memory 430, and a communication bus 440, wherein the processor 410, the communication interface 420, and the memory 430 communicate with each other via the communication bus 440. The processor 410 can call logical instructions in the memory 430 to execute a macro-micro traffic control method based on an optimization solver, including: Acquire multi-source traffic data for the target area; wherein the multi-source traffic data includes at least one of historical travel data, population employment attributes, and origin-destination data; The traffic control scheme generation steps are as follows: A macro-level model is invoked to determine the origin-destination demand matrix based on the multi-source traffic data; the origin-destination demand matrix is ​​input into a meso-level model, and a dynamic traffic assignment algorithm is used to generate traffic flow parameters; a mixed-integer programming model is constructed based on the traffic flow parameters, and the mixed-integer programming model is solved based on a preset solver strategy to obtain the traffic control scheme; wherein, the traffic control scheme includes a signal control scheme and a vehicle scheduling scheme; the preset solver strategy is determined based on the historical performance of each solver in solving the mixed-integer programming model; The optimal road network state determination steps are as follows: based on the traffic control scheme, the road network operation state is updated, the deviation rate between the updated road network state and the real-time collected traffic monitoring data is determined, and if the deviation rate does not exceed a preset threshold, the current road network state is determined as the optimal road network state.

[0098] Furthermore, the logical instructions in the aforementioned memory 430 can be implemented as software functional units and, when sold or used as independent products, can be stored in a computer-readable storage medium. Based on this understanding, the technical solution of this application, in essence, or the part that contributes to the prior art, or a part of the technical solution, can be embodied in the form of a software product. This computer software product is stored in a storage medium and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute all or part of the steps of the methods described in the various embodiments of this application. The aforementioned storage medium includes various media capable of storing program code, such as USB flash drives, portable hard drives, read-only memory (ROM), random access memory (RAM), magnetic disks, or optical disks.

[0099] On the other hand, this application also provides a computer program product, which includes a computer program that can be stored on a non-transitory computer-readable storage medium. When the computer program is executed by a processor, the computer is able to execute a macro-micro traffic control method based on an optimization solver provided by the above methods, including: Acquire multi-source traffic data for the target area; wherein the multi-source traffic data includes at least one of historical travel data, population employment attributes, and origin-destination data; The traffic control scheme generation steps are as follows: A macro-level model is invoked to determine the origin-destination demand matrix based on the multi-source traffic data; the origin-destination demand matrix is ​​input into a meso-level model, and a dynamic traffic assignment algorithm is used to generate traffic flow parameters; a mixed-integer programming model is constructed based on the traffic flow parameters, and the mixed-integer programming model is solved based on a preset solver strategy to obtain the traffic control scheme; wherein, the traffic control scheme includes a signal control scheme and a vehicle scheduling scheme; the preset solver strategy is determined based on the historical performance of each solver in solving the mixed-integer programming model; The optimal road network state determination steps are as follows: based on the traffic control scheme, the road network operation state is updated, the deviation rate between the updated road network state and the real-time collected traffic monitoring data is determined, and if the deviation rate does not exceed a preset threshold, the current road network state is determined as the optimal road network state.

[0100] Furthermore, this application also provides a non-transitory computer-readable storage medium storing a computer program thereon, which, when executed by a processor, is implemented to perform a macro-micro traffic control method based on an optimization solver provided by the methods described above, comprising: Acquire multi-source traffic data for the target area; wherein the multi-source traffic data includes at least one of historical travel data, population employment attributes, and origin-destination data; The traffic control scheme generation steps are as follows: A macro-level model is invoked to determine the origin-destination demand matrix based on the multi-source traffic data; the origin-destination demand matrix is ​​input into a meso-level model, and a dynamic traffic assignment algorithm is used to generate traffic flow parameters; a mixed-integer programming model is constructed based on the traffic flow parameters, and the mixed-integer programming model is solved based on a preset solver strategy to obtain the traffic control scheme; wherein, the traffic control scheme includes a signal control scheme and a vehicle scheduling scheme; the preset solver strategy is determined based on the historical performance of each solver in solving the mixed-integer programming model; The optimal road network state determination steps are as follows: based on the traffic control scheme, the road network operation state is updated, the deviation rate between the updated road network state and the real-time collected traffic monitoring data is determined, and if the deviation rate does not exceed a preset threshold, the current road network state is determined as the optimal road network state.

[0101] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0102] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, or network device, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0103] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A macro-micro traffic control method based on an optimization solver, characterized in that, include: Acquire multi-source traffic data for the target area; wherein the multi-source traffic data includes at least one of historical travel data, population employment attributes, and origin-destination data; The traffic control scheme generation steps are as follows: A macro-level model is invoked to determine the origin-destination demand matrix based on the multi-source traffic data; the origin-destination demand matrix is ​​input into a meso-level model, and a dynamic traffic assignment algorithm is used to generate traffic flow parameters; the traffic flow parameters are used as input conditions for a micro-level model, and a mixed-integer programming model is constructed based on the traffic flow parameters; the mixed-integer programming model is solved based on a preset solver strategy to obtain a traffic control scheme; wherein, the traffic control scheme includes a signal control scheme and a vehicle scheduling scheme; the preset solver strategy is determined based on the historical performance of each solver in solving the mixed-integer programming model; the mixed-integer programming model is constructed based on the traffic flow parameters... The traffic control scheme is obtained by solving the mixed-integer programming model using a preset solver strategy, and further includes: using the passenger set, station set, and vehicle set as second decision variables; constructing a second objective function based on the fixed cost of vehicles and the variable cost of vehicles serving passengers at stations; constructing a second mixed-integer programming model based on the second decision variables and the second objective function, wherein the constraints of the second mixed-integer programming model include capacity constraints, time window constraints, and service continuity constraints; and solving the second mixed-integer programming model using a preset solver strategy to generate a vehicle scheduling scheme; wherein the vehicle scheduling scheme includes a vehicle departure plan, a stop sequence, and a passenger allocation scheme. The optimal road network state determination steps are as follows: based on the traffic control scheme, the road network operation state is updated, the deviation rate between the updated road network state and the real-time collected traffic monitoring data is determined, and if the deviation rate does not exceed a preset threshold, the current road network state is determined as the optimal road network state.

2. The macro-micro traffic control method based on an optimization solver as described in claim 1, characterized in that, The step of determining the origin-destination demand matrix based on the multi-source traffic data includes: The target area is divided into traffic zones, resulting in multiple traffic zones. Establish the spatial mapping relationship between the traffic zones and road network nodes; Based on the historical travel data, population employment attributes, and origin-destination data of the aforementioned traffic zones, the origin-destination demand matrix of each traffic zone is predicted. Based on the spatial mapping relationship, the origin-destination demand matrix of each traffic zone is converted into the origin-destination demand matrix between the corresponding road network nodes.

3. The macro-micro traffic control method based on an optimization solver as described in claim 1, characterized in that, The process of constructing a mixed-integer programming model based on the traffic flow parameters, solving the mixed-integer programming model based on a preset solver strategy, and obtaining a traffic control scheme includes: The signal cycle length and the green light duration for each phase are used as the first decision variables; Based on the traffic flow parameters, a first objective function is constructed with the goal of minimizing the average vehicle delay at the intersection. The constraints of the first objective function include periodic constraints, green light duration constraints, total number of green lights constraints, and phase sequence constraints. Construct a first mixed-integer programming model based on the first decision variable and the first objective function; The first mixed integer programming model is solved using a preset solver strategy to obtain a signal control scheme.

4. The macro-micro traffic control method based on an optimization solver as described in claim 1, characterized in that, The determination of the preset solver strategy includes: Determine the task type of the mixed integer programming model, and query the historical performance of each solver based on the task type; The overall score for each solver is determined based on its historical performance. Select the preset quantity solver with the highest comprehensive score.

5. The macro-micro traffic control method based on an optimization solver as described in claim 1, characterized in that, The updated road network status includes the average vehicle speed of road segments, and determining the deviation rate between the updated road network status and the real-time collected traffic monitoring data includes: Real-time data collection of average vehicle speed and the number of monitored road segments; The deviation rate is determined based on the average vehicle speed, the number of monitored road segments, and the average vehicle speed of the road segments.

6. The macro-micro traffic control method based on an optimization solver as described in claim 1, characterized in that, The step of updating the road network operating status based on the traffic control scheme and determining the deviation rate between the updated road network status and the real-time collected traffic monitoring data further includes: If the deviation rate exceeds a preset threshold, the model parameters of the mesoscopic model are adjusted, and the traffic control scheme generation step and the optimal road network state determination step are repeated.

7. A macro-micro traffic control device based on an optimization solver, characterized in that, include: A multi-source traffic data acquisition module is used to acquire multi-source traffic data of a target area; wherein, the multi-source traffic data includes at least one of historical travel data, population employment attributes, and origin-destination data; A traffic control scheme generation module is used to call a macro-level model to determine the origin-destination demand matrix based on the multi-source traffic data; input the origin-destination demand matrix into a meso-level model, and generate traffic flow parameters using a dynamic traffic assignment algorithm; construct a mixed-integer programming model based on the traffic flow parameters, and solve the mixed-integer programming model based on a preset solver strategy to obtain a traffic control scheme; wherein, the traffic control scheme includes a signal control scheme and a vehicle scheduling scheme; the preset solver strategy is determined based on the historical performance of each solver in solving the mixed-integer programming model; the construction of the mixed-integer programming model based on the traffic flow parameters and the solution of the mixed-integer programming model based on the preset solver strategy are described in detail. Solving the mixed-integer programming model to obtain a traffic control scheme further includes: using the passenger set, station set, and vehicle set as second decision variables; constructing a second objective function based on the fixed cost of vehicles and the variable cost of vehicles serving passengers at stations; constructing a second mixed-integer programming model based on the second decision variables and the second objective function, wherein the constraints of the second mixed-integer programming model include capacity constraints, time window constraints, and service continuity constraints; solving the second mixed-integer programming model using a preset solver strategy to generate a vehicle scheduling scheme; wherein the vehicle scheduling scheme includes a vehicle departure plan, a stop sequence, and a passenger allocation scheme. The optimal road network state determination module is used to update the road network operation state based on the traffic control scheme, determine the deviation rate between the updated road network state and the real-time collected traffic monitoring data, and determine the current road network state as the optimal road network state if the deviation rate does not exceed a preset threshold.

8. An electronic device comprising a memory, a processor, and a computer program stored in the memory and running on the processor, characterized in that, When the processor executes the computer program, it implements the macro-micro traffic control method based on an optimization solver as described in any one of claims 1 to 6.

9. A non-transitory computer-readable storage medium having a computer program stored thereon, characterized in that, When the computer program is executed by the processor, it implements the macro-micro traffic control method based on the optimization solver as described in any one of claims 1 to 6.

Images