Highway traffic flow resilience regulation method and system under complex construction scene

Abstract

This invention discloses a method and system for resilient traffic flow control on highways under complex construction scenarios, relating to the field of traffic control. The method includes: zoning and calibrating the construction area and traffic sensing points to determine the road network construction topology; sampling and predicting real-time traffic flow data based on the traffic sensing network, and determining traffic congestion waves through wave eigenvector transformation; constructing resilient control units based on the road network construction topology and deploying them at the road network edge; intervening in the traffic congestion waves using wave interference offsetting and detour diversion, and then iteratively allocating the solutions to each zone using a trust region framework to determine the resilient control scheme; and performing dynamic control and management of traffic flow resilience. This application solves the technical problem of existing dynamic traffic flow control methods, which struggle to quantify and identify congestion evolution patterns in real time and implement precise and efficient interventions, achieving the technical effect of real-time quantification and identification of congestion evolution patterns and precise and efficient interventions for traffic flow.

Description

Technical Field

[0001] This application relates to the field of traffic control, and in particular to methods and systems for regulating the resilience of highway traffic flow under complex construction scenarios. Background Technology

[0002] Traffic flow stability management in highway construction areas directly affects road traffic efficiency, driving safety, and construction order, making it a critical issue that must be addressed in highway operation and construction. Currently, traffic flow control in construction sections mainly relies on fixed construction closure schemes combined with manual on-site guidance, conventional traffic lights and information boards, and simple route diversion. These methods depend on static preset parameters and cannot match the dynamic evolution of traffic flow and changes in construction boundaries in real time. They also struggle to quantitatively identify and accurately intervene in the congestion propagation process, easily leading to problems such as continued congestion spread, delayed traffic management, and low diversion efficiency.

[0003] At present, the relevant technologies face the technical challenge of making it difficult to quantify and identify the evolution of congestion patterns in real time and implement precise and efficient interventions in the dynamic control of highway traffic flow under complex construction scenarios. Summary of the Invention

[0004] This application provides a method and system for resilient traffic flow control on highways under complex construction scenarios. It employs a method of zoning and calibrating construction units and traffic sensing points within the road network to construct a road network construction topology. Real-time traffic flow data is sampled and predicted using a traffic sensing network. Traffic congestion waves are obtained through wave eigenvector transformation. Based on the road network construction topology, a resilient control unit is constructed at the road network edge, consisting of a first intervention component of the resilient capacity spectrum and a second equilibrium component of the zoning coupling matrix. The traffic congestion waves are then addressed through wave interference offsetting and detour diversion interventions. Through iterative allocation of control tasks to each zone using a trust region framework, the control task is transformed into a resilient control scheme. The traffic management system responds to this scheme, enabling dynamic control and management of highway traffic flow resilience. This approach solves the technical problem of difficulty in real-time quantification and identification of congestion evolution patterns and the implementation of precise and efficient interventions in dynamic control of highway traffic flow under complex construction scenarios. It achieves the technical effect of real-time quantification and identification of congestion evolution patterns and precise and efficient intervention of traffic flow.

[0005] This application provides a method for resilient traffic flow control on highways under complex construction scenarios, comprising: performing zone calibration and traffic sensing point calibration of the construction area in the traffic network based on construction units to determine the road network construction topology; performing sampling and prediction of real-time traffic flow data based on the traffic sensing network composed of traffic sensing points, and determining traffic congestion waves by performing wave feature vector transformation; constructing a resilience control unit based on the road network construction topology and deploying it on the edge of the road network, and solving the traffic congestion waves through intervention based on wave interference offsetting and detour diversion, and allocating the solution to each zone through iterative solution using a trust region framework, thereby determining a resilience control scheme, wherein the resilience control unit is composed of a first intervention component based on a resilience capacity spectrum and a second equilibrium component based on a zone coupling matrix cascaded; and the traffic management system responds to the resilience control scheme to perform dynamic control and management of traffic flow resilience.

[0006] In a possible implementation, real-time traffic flow data is sampled and predicted. Traffic congestion waves are determined by performing wave feature vector conversion. The following processing is performed: real-time flow and direction prediction for the real-time traffic flow data in a preset time zone is performed based on the lightweight prediction unit to determine traffic flow prediction data; traffic evolution trajectories with multiple traffic feature threads are generated based on the real-time traffic flow data and traffic flow prediction data; and traffic congestion waves are determined based on the traffic evolution trajectories by performing wave feature vector conversion.

[0007] In a possible implementation, the following processing is performed: identifying the traffic evolution trajectory, determining waveform elements, wherein the amplitude is defined based on the difference between congestion density and free flow density, the wave speed is defined based on the speed at which the congestion wave peak propagates upstream, the wavelength is defined based on the product of the congestion duration and the wave speed, and the phase is defined based on the position of the congestion wave peak; and the traffic evolution trajectory is reconstructed into the traffic congestion wave based on the waveform elements.

[0008] In a possible implementation, before building the resilience control unit, the following processing is performed: for each construction unit calibrated by the partition, a capacity-boundary response surface is constructed according to the engineering characteristics; Gaussian process regression learning is performed based on the response surface to construct a resilience capacity spectrum, wherein the output of the resilience capacity spectrum is the distribution of the passage capacity of the partition at different boundary locations.

[0009] In a possible implementation, a resilience control unit is built based on the road network construction topology, and the following processing is performed: a partition coupling matrix is ​​established according to the road network construction topology, wherein the diagonal elements are determined by the influence intensity of the change of the first partition boundary on the partition flow, and the off-diagonal elements are determined by the coupling transfer entropy of the influence of the change of the first partition boundary on the upstream and downstream adjacent partitions; a first intervention component is built according to the resilience capacity spectrum, a second equilibrium component is built according to the partition coupling matrix, and the first intervention component and the second equilibrium component are cascaded to form the resilience control unit.

[0010] In a possible implementation, a resilience control scheme is determined, and the following processes are performed: the resilience control unit receives the traffic congestion wave and initializes the road network construction topology; according to the first intervention component, an intervention solution based on wave interference offsetting and detour diversion is performed to determine the intervention wave parameters, wherein the intervention wave parameters include offsetting intervention waveform elements, diversion intervention waveform elements, and intervention timing; according to the second equalization component, a trust region framework iterative solution based on the single-zone intervention allocation of the intervention wave parameters is performed to determine the zone intervention wave parameters; the zone intervention wave parameters are converted into boundary adjustment sequences and detour parameters to determine the resilience control scheme.

[0011] In a possible implementation, after determining the resilience control scheme, the following processing is performed: according to the resilience control scheme, a digital control simulation is performed in the road network construction topology to determine the control simulation data; abnormal traffic flow characteristics in the control simulation data are identified to generate local intervention conditions; according to the local intervention conditions, an optimization solution based on the local intervention conditions is performed to optimize the resilience control scheme.

[0012] In a possible implementation, the traffic management system responds to the resilience control scheme, performs dynamic control and management of traffic flow resilience, and executes the following processes: establishing communication interaction between the road network edge and the traffic management system; the traffic management system receives the resilience control scheme, generates a first traffic flow scheduling instruction and a second construction control map; controls traffic facilities according to the first traffic flow scheduling instruction, and distributes the second construction control map to the vehicle platform port for real-time construction boundary prompts.

[0013] In a possible implementation, after dynamic regulation and management of traffic flow resilience, the following processing is performed: the traffic sensing network continuously monitors traffic flow data, performs differential comparison of traffic congestion waves before and after regulation, and evaluates the regulation effect of each zone, wherein the amplitude attenuation rate, wave velocity change and queue length reduction are used as quantitative evaluation standards; and the resilience regulation unit is updated and learned according to the regulation effect and the progress of zone engineering.

[0014] This application also provides a highway traffic flow resilience control system for complex construction scenarios, including: a road network construction topology determination module, used to perform zoning and traffic sensing point calibration of construction areas in the traffic network based on construction units, and determine the road network construction topology; a traffic congestion wave determination module, used to perform sampling and prediction of real-time traffic flow data based on the traffic sensing network composed of traffic sensing points, and determine the traffic congestion wave by performing wave feature vector conversion; a resilience control module, used to build a resilience control unit based on the road network construction topology, deployed on the edge side of the road network, and solve the traffic congestion wave by intervention based on wave interference offsetting and detour diversion, and distribute it to each zone through iterative solution using a trust region framework, and convert to determine the resilience control scheme, wherein the resilience control unit is composed of a first intervention component based on the resilience capacity spectrum and a second equilibrium component based on the zoning coupling matrix; and a dynamic control management module, used to respond to the resilience control scheme through the traffic management system and perform dynamic control management of traffic flow resilience.

[0015] This application proposes a method and system for resilient traffic flow control in complex construction scenarios on highways. First, the construction area is zoned and calibrated based on construction units within the road network, and traffic sensing points are identified to determine the road network construction topology. Then, based on the traffic sensing network formed by the traffic sensing points, real-time traffic flow data is sampled and predicted. Traffic congestion waves are determined through wave eigenvector transformation. Next, a resilient control unit is constructed based on the road network construction topology and deployed at the road network edge. The traffic congestion waves are addressed through intervention solutions based on wave interference offsetting and detour diversion. These solutions are then iteratively distributed to each zone using a trust region framework, and a resilient control scheme is determined. The resilient control unit is constructed by cascading a first intervention component based on a resilience capacity spectrum and a second equilibrium component based on a zone coupling matrix. Finally, the traffic management system responds to the resilient control scheme, dynamically managing traffic flow resilience. Through this process, the method and system proposed in this application achieve the technical effect of real-time quantitative identification of congestion evolution patterns and precise and efficient intervention in traffic flow. Attached Figure Description

[0016] To more clearly illustrate the technical solutions of the embodiments of the present invention, the accompanying drawings of the embodiments of the present invention will be briefly described below. Flowcharts are used in this application to illustrate the operations performed by the system according to the embodiments of the present application. It should be understood that the preceding or following operations are not necessarily performed precisely in sequence. Instead, various steps can be processed in reverse order or simultaneously as needed. Furthermore, other operations can be added to these processes, or one or more steps can be removed from these processes.

[0017] Figure 1 A flowchart illustrating the method for regulating the resilience of highway traffic flow under complex construction scenarios provided in this application embodiment.

[0018] Figure 2 A schematic diagram of the structure of a highway traffic flow resilience control system under complex construction scenarios provided in this application embodiment.

[0019] Explanation of reference numerals in the attached diagram: Module 10 for determining the road network construction topology, Module 20 for determining traffic congestion waves, Module 30 for resilience regulation, and Module 40 for dynamic regulation and management. Detailed Implementation

[0020] To further illustrate the technical means and effects of the present invention in achieving its intended purpose, the following detailed description of the specific implementation methods, structures, features, and effects of the present invention, in conjunction with the accompanying drawings and preferred embodiments, is provided below.

[0021] This application provides a method for regulating the resilience of highway traffic flow under complex construction scenarios, such as... Figure 1 As shown, the method includes: Step S100: In the traffic network, the construction area is zoned and calibrated based on construction units and traffic sensing points to determine the construction topology of the road network.

[0022] Specifically, within the road network affected by highway construction, the construction sections are divided into independent construction units based on construction segments, work surfaces, and closed areas. Cameras, radar, and geomagnetic coils are deployed at key locations as traffic sensing points, ultimately forming a road network construction topology with a construction area structure, clearly defining where construction is underway, where traffic is open, and where data is collected. Specifically, a highway electronic map vector layer is overlaid with construction CAD drawings. Construction units are divided according to the principles of closed length not exceeding 500 meters, uniform work type, and clear upstream and downstream connections. Each unit is numbered and its starting and ending chainages, number of closed lanes, and temporary opening locations are recorded. Traffic sensing points are deployed at a density of one every 500 meters, with denser deployment at the start, end, divergence, and merging points of the construction area. Millimeter-wave traffic radar, video traffic detectors, and geomagnetic vehicle detectors are used. The coordinates, types, coverage areas, and associated construction units of the sensing points are entered into the topology database, ultimately generating a road network construction topology containing nodes, road segments, construction units, sensing points, traffic directions, and traffic capacity constraints.

[0023] Step S200: Based on the traffic sensing network formed by traffic sensing points, perform sampling and prediction of real-time traffic flow data, and determine the traffic congestion wave by performing wave feature vector conversion.

[0024] Specifically, traffic sensing points collect real-time data such as traffic flow, speed, and density to predict short-term traffic conditions. Traffic changes are then transformed into wave-like traffic congestion waves, with amplitude, velocity, wavelength, and phase used to describe the intensity, propagation speed, impact range, and location of the congestion. Specifically, traffic flow, speed, density, time occupancy, and queue length are acquired from traffic sensing points once per second. Lightweight prediction units perform rolling predictions of future traffic flow, stitching together multi-dimensional traffic features to form a traffic evolution trajectory. This trajectory is then transformed into four characteristic quantities—amplitude, velocity, wavelength, and phase—of the congestion wave through waveform transformation.

[0025] In one possible implementation, real-time traffic flow data is sampled and predicted. Traffic congestion waves are determined through wave feature vector transformation. Step S200 further includes step S210, whereby a lightweight prediction unit performs real-time flow and direction prediction on the real-time traffic flow data for a preset time zone to determine traffic flow prediction data. Specifically, a lightweight prediction unit consisting of a one-dimensional convolutional layer, a gated recurrent unit layer, and a fully connected output layer is used. Historical real-time traffic flow data within a preset time period is used as input. The input data includes five traffic features: flow rate, speed, density, time occupancy, and queue length. Sliding window sampling is performed at fixed time intervals to predict the cross-sectional traffic flow and the proportion of straight, left-turn, and right-turn directions for each lane in future time periods within the preset time zone. The output is traffic flow prediction data aligned with the spatiotemporal dimension of the real-time traffic flow data. The prediction data also includes future changes in the five features: flow rate, speed, density, time occupancy, and queue length.

[0026] Step S220: Based on the real-time traffic flow data and traffic flow prediction data, a traffic evolution trajectory with multiple traffic feature threads is generated. Specifically, the traffic flow prediction data output in S210 and the corresponding real-time traffic flow data are aligned and continuously spliced ​​according to the same timestamp, and five independent and spatiotemporally synchronized traffic feature threads are constructed with a unified time axis. These are the flow rate evolution trajectory, speed evolution trajectory, density evolution trajectory, time occupancy evolution trajectory, and queue length evolution trajectory. Each trajectory includes historical measured data segments and future predicted data segments. The five trajectories together constitute a set of multi-feature traffic evolution trajectories for congestion wave identification.

[0027] Step S230: Based on the traffic evolution trajectory, perform wave feature vector conversion to determine the traffic congestion wave. Specifically, the five traffic evolution trajectories generated in S220—flow rate, speed, density, time occupancy, and queue length—are uniformly subjected to moving average filtering for noise reduction. The density evolution trajectory is used as the main feature for waveform element extraction, while the four trajectories—flow rate, speed, time occupancy, and queue length—are used as auxiliary verification and joint judgment features. Through multi-trajectory collaborative identification of traffic state inflection points and congestion peak points, the continuous traffic evolution trajectory is converted into a wave feature vector containing amplitude, wave speed, wavelength, and phase, ultimately determining the traffic congestion wave that can be used for wave interferometry offset calculation.

[0028] In one possible implementation, step S230 further includes step S231, identifying the traffic evolution trajectory and determining waveform elements. Specifically, the amplitude is defined based on the difference between congestion density and free-flow density; the wave speed is defined based on the speed at which the congestion peak propagates upstream; the wavelength is defined based on the product of the congestion duration and the wave speed; and the phase is defined based on the position of the congestion peak. Specifically, the five traffic evolution trajectories generated in S220—flow rate, speed, density, time occupancy, and queue length—are filtered and denoised. Then, the free-flow state interval and the congestion state interval are jointly determined using the five trajectories. The free-flow state interval refers to a stable state interval where the speed trajectory is above the highway free-flow threshold, the density trajectory and time occupancy trajectory are at a low level, and the queue length trajectory remains at zero. The congestion state interval refers to a stable state interval where the speed trajectory is below the congestion determination threshold, the density trajectory and time occupancy trajectory are significantly increased, and the queue length trajectory continues to grow. The density trajectory is considered to fall within the congestion state interval and the free-flow state interval. The amplitude is determined by the average difference of the values. The arrival time difference of the congestion peak at multiple continuous sections upstream of the construction area is identified by using the density trajectory and the queue length trajectory. The wave speed of the congestion wave propagating upstream is calculated by combining the actual spatial distance between adjacent sensing sections. The congestion start time and congestion dissipation time are confirmed by using the density trajectory, velocity trajectory, and time occupancy trajectory. The difference between the two times is taken as the congestion duration and multiplied by the absolute value of the wave speed to obtain the wavelength. The phase of the congestion wave peak is determined by combining the road station position corresponding to the density peak with the distribution position of the queue length peak. Thus, the four waveform elements of amplitude, wave speed, wavelength, and phase are determined.

[0029] Step S232: Based on the waveform elements, the traffic evolution trajectory is reconstructed into the traffic congestion wave. Specifically, a congestion waveform model is constructed by superimposing a time decay term and a spatial decay term onto a sinusoidal function. The wave amplitude, wave velocity, wavelength, and phase obtained in S231 are used as model inputs. The five traffic evolution trajectories generated in S220—flow rate, speed, density, time occupancy rate, and queue length—are used as spatiotemporal constraints to fit and correct the waveform model, outputting a congestion wave function that continuously varies with time and space. This congestion wave propagates along the upstream direction of the highway and is used for wave interference offset calculation and intervention wave cancellation effect determination in the resilience control unit.

[0030] Step S300: Based on the road network construction topology, a resilience control unit is built and deployed on the edge of the road network. The traffic congestion wave is solved by intervention based on wave interference offsetting and detour diversion. The solution is distributed to each zone through iterative solution using a trust region framework. The resilience control scheme is then determined by transformation. The resilience control unit is composed of a first intervention component based on the resilience capacity spectrum and a second equalization component based on the zone coupling matrix.

[0031] Specifically, the road network construction topology is used as the basic structure, and resilience control units are deployed at the roadside edge computing nodes. Each unit consists of a first intervention component based on the resilience capacity spectrum and a second equalization component based on the partition coupling matrix. The amplitude, velocity, wavelength, and phase of the traffic congestion wave are used as inputs. First, the intervention objectives of wave interference offsetting and detour diversion are solved through the first intervention component. Then, under the condition of satisfying the traffic capacity constraints and partition coupling influence constraints of each construction zone, the second equalization component uses a trust region framework to perform iterative optimization calculations, decompose and distribute the overall intervention amount to each construction zone, and finally convert the iterative optimization results into executable parameters such as construction boundary adjustment, lane control, and diversion guidance, forming a highway traffic flow resilience control scheme for complex construction scenarios.

[0032] In one possible implementation, before constructing the resilience control unit, step S300 further includes step S310, which involves constructing a capacity-boundary response surface for each construction unit calibrated by the zone, based on the engineering characteristics. Specifically, the construction boundary offset, the number of closed lanes, the number and location of temporary openings, and the space occupied by the operating machinery for each construction unit are used as input variables, and the actual maximum traffic capacity of the construction unit under the corresponding state is used as the output variable. By combining on-site measured data with theoretical calculations of highway traffic capacity, multiple sets of traffic capacity samples under different construction boundary conditions are collected. A three-dimensional interpolation method is used to construct a continuous capacity-boundary response surface with the construction boundary as the independent variable and traffic capacity as the dependent variable, establishing a quantitative correspondence between changes in the construction boundary and the actual traffic capacity.

[0033] Step S320: Based on the response surface, Gaussian process regression learning is performed to construct a resilience capacity spectrum, wherein the output of the resilience capacity spectrum is the traffic capacity distribution of the partition at different boundary locations. Specifically, the capacity-boundary response surface samples constructed in S310 are used as the training dataset, and a Gaussian process regression model is used for supervised learning. The model takes the construction unit number, current construction progress, construction boundary location, and lane closure status as input features, and takes the expected traffic capacity, traffic capacity fluctuation range, and reliable traffic capacity lower limit under the corresponding conditions as outputs. The kernel function parameters and noise variance of the model are determined through training. After learning is completed, an online queryable resilience capacity spectrum is formed, which can quickly output the corresponding traffic capacity distribution results for any construction boundary conditions.

[0034] In one possible implementation, a resilience control unit is built based on the road network construction topology. Step S300 further includes step S330, which establishes a partition coupling matrix based on the road network construction topology. The diagonal elements are determined by the intensity of the impact of the first partition boundary change on the partition traffic flow, and the off-diagonal elements are determined by the coupling transfer entropy of the impact of the first partition boundary change on upstream and downstream adjacent partitions. Specifically, a partition coupling matrix of the same dimension is constructed using the number of construction partitions in the road network construction topology as the matrix dimension. The diagonal elements represent the direct impact intensity of the current construction partition's boundary change on the traffic flow and traffic status of that partition, calculated by the ratio of the construction partition boundary adjustment to the corresponding capacity change. The off-diagonal elements represent the coupling transfer impact intensity of a construction partition boundary change on the traffic status of upstream and downstream adjacent construction partitions, calculated using the transfer entropy method to calculate the degree of influence between the traffic flow time series of adjacent partitions and normalized as the off-diagonal element value. Finally, a partition coupling matrix that fully reflects the traffic association characteristics between partitions is formed.

[0035] Step S340: Construct a first intervention component based on the resilience capacity spectrum and a second equalization component based on the partition coupling matrix. Cascade the first intervention component and the second equalization component to form the resilience control unit. Specifically, the resilience capacity spectrum constructed in S320 is encapsulated into a first intervention component with real-time traffic capacity query and intervention threshold determination functions. The partition coupling matrix constructed in S330 is encapsulated into a second equalization component with partition traffic impact calculation and traffic flow equalization allocation functions. The output intervention target, intervention waveform, and total intervention amount of the first intervention component are used as input conditions for the second equalization component. Cascade the components in a manner where the first intervention component precedes the second equalization component, forming a complete resilience control unit with traffic congestion waves as input and partition-level control parameters as output. This unit is then deployed to a road network edge computing device for operation.

[0036] In one possible implementation, determining the resilience control scheme, step S300 further includes step S350, whereby the resilience control unit receives the traffic congestion wave and initializes the road network construction topology with data. Specifically, the resilience control unit receives the traffic congestion wave data output in S232, including four waveform parameters: amplitude, wave velocity, wavelength, and phase. Simultaneously, it reads the construction zone information, sensing point locations, lane configurations, current construction progress, and real-time traffic status data from the road network construction topology. It then sets the initial trust region radius, maximum number of iterations, convergence accuracy threshold, and construction safety constraint boundary of the trust region optimization framework, completing the initialization of all data and parameters before the control calculation.

[0037] Step S360: Based on the first intervention component, perform intervention solution based on wave interference offsetting and detour diversion to determine the intervention wave parameters. These parameters include offsetting intervention waveform elements, diversion intervention waveform elements, and intervention timing. Specifically, the first intervention component takes the traffic congestion wave as the object, with the optimization objectives of minimizing congestion wave energy and shortening queue length. Based on the waveform interference principle, it obtains an offsetting intervention waveform with opposite phase and matching amplitude to the congestion wave. Simultaneously, it calculates the detour diversion starting point, diversion ratio, and detour path capacity by combining the road network topology and resilience capacity spectrum. Finally, it determines the complete intervention wave parameters, including offsetting intervention waveform elements, diversion intervention waveform elements, and the optimal intervention initiation timing.

[0038] Step S370: Based on the second equalization component, perform iterative solution of the trust region framework under single-zone intervention allocation based on the intervention wave parameters to determine the zone intervention wave parameters. Specifically, the second equalization component uses the intervention wave parameters obtained in S360 as the overall control target, the zone coupling matrix as the traffic impact constraint between zones, and the capacity boundaries of each zone provided by the resilience capacity spectrum as the feasible region constraint. Within the trust region optimization framework, it iteratively adjusts the intervention wave amplitude, intervention intensity, and intervention timing of each construction zone. In each iteration, the objective function value and the degree of constraint satisfaction are calculated, and the trust region radius is dynamically updated until the convergence accuracy requirement is met. Finally, the independent zone intervention wave parameters allocated to each construction zone are output.

[0039] Step S380 involves converting the boundary adjustment sequence and detour parameters of the partition intervention wave parameters to determine the resilience control scheme. Specifically, the partition intervention wave parameters obtained in S370 are converted into the construction boundary offset, number of lanes open and closed, variable speed limit, turning ratio at diversion intersections, and detour route selection scheme of each construction partition at different times according to the mapping relationship between waveform parameters and traffic control measures. This forms a set of digital control instructions arranged in time sequence and divided by partition. This set of instructions is the highway traffic flow resilience control scheme that can be directly issued to the traffic management system for execution.

[0040] In one possible implementation, after determining the resilience control scheme, step S300 further includes step S390, which involves performing a digital control simulation on the road network construction topology according to the resilience control scheme to determine the control simulation data. Specifically, the road network construction topology, real-time traffic flow data, construction unit parameters, and the resilience control scheme obtained in S380 are input into the microscopic traffic simulation engine. The simulation is executed step by step according to the time-sequential construction boundary adjustments and traffic diversion instructions in the scheme, with a fixed step size to advance the simulation process. Throughout the process, traffic state data such as flow rate, speed, density, queue length, and travel time are collected from each construction zone and key section to form complete control simulation data.

[0041] Step S3100: Identify abnormal traffic flow characteristics in the control simulation data and generate local intervention conditions. Specifically, the control simulation data obtained in S390 is traversed and detected time-by-time and zone-by-zone. Anomaly judgment thresholds are set, including the duration of speed below the set threshold, queue length exceeding the safety limit, sudden change in traffic flow at adjacent cross-sections, and increase in road segment travel time. When a traffic flow characteristic that meets any of the thresholds appears in the simulation data, it is judged as an abnormal traffic state, and the corresponding abnormal zone and abnormal time period are located. Local intervention conditions containing abnormal location, abnormal type, and constraint correction values ​​are generated.

[0042] Step S3110: Based on the local intervention conditions, perform optimization based on the local intervention conditions to optimize the resilience control scheme. Specifically, the local intervention conditions generated in S3100 are added to the constraint set of the trust region framework, keeping the original optimization objective unchanged, and the iterative optimization process is restarted. The intervention wave parameters, construction boundary adjustment amount, diversion timing and diversion ratio of the corresponding construction zone are adjusted in a targeted manner to eliminate abnormal traffic flow characteristics that appear during the simulation. After optimization, a final version of the resilience control scheme that is convergent, stable and without obvious anomalies is formed.

[0043] In one possible implementation, after dynamic regulation and management of traffic flow resilience, step S300 further includes step S3120, whereby the traffic sensing network continuously monitors traffic flow data, performs differential comparison of traffic congestion waves before and after regulation, and evaluates the regulation effect of each zone. The amplitude attenuation rate, wave velocity change, and queue length reduction are used as quantitative evaluation standards. Specifically, the traffic sensing network continuously collects data on flow rate, speed, density, time occupancy, and queue length in real time. Following the same method as in S230 to S232, it re-identifies the regulated traffic congestion waves, performs differential comparison between the regulated congestion waves and the original congestion waves before regulation, and calculates three quantitative indicators: amplitude attenuation rate, wave velocity change, and queue length reduction. The amplitude attenuation rate is the ratio of the amplitude difference before and after regulation to the amplitude before regulation; the wave velocity change is the difference between the absolute value of the wave velocity after regulation and the absolute value of the wave velocity before regulation; and the queue length reduction is the difference between the maximum queue length before and after regulation. These three indicators are used to comprehensively evaluate the traffic flow resilience regulation effect of each construction zone.

[0044] Step S3130: Based on the control effect and the progress of the zoned project, the resilience control unit is updated and learned. Specifically, the quantitative indicators of the control effect of each zone obtained in S3120, the current actual project progress of the zone, and the latest traffic flow and construction boundary data are used as incremental learning samples and input into the resilience control unit. The Gaussian process regression model, the zone coupling matrix, and the trust region optimization parameters are updated and fine-tuned online. The model iteration is completed without interrupting the real-time control service, so that the resilience capacity spectrum and the zone coupling relationship are more in line with the current construction and traffic conditions, improving the accuracy and adaptability of subsequent traffic congestion wave intervention and diversion control.

[0045] In step S400, the traffic management system responds to the resilience control scheme and performs dynamic control and management of traffic flow resilience.

[0046] Specifically, the traffic management system receives the final resilience control scheme generated by S380 and S3110 through the communication link, analyzes the zonal control instructions, timing execution plans, traffic control parameters and construction boundary adjustment requirements in the scheme, and coordinates with roadside traffic facilities, vehicle navigation terminals and construction control equipment to synchronously and dynamically control traffic flow and construction areas according to preset timing, so as to achieve coordinated execution of congestion wave interference offsetting and vehicle detour diversion.

[0047] In one possible implementation, the traffic management system responds to the resilience control scheme by dynamically controlling traffic flow resilience. Step S400 further includes step S410, establishing communication interaction between the road network edge and the traffic management system. Specifically, a dual-link structure with wired fiber optic and 5G wireless communication as backups is adopted. A real-time communication connection is established between the road network edge resilience control unit and the central traffic management system based on a lightweight message transmission protocol. Periodic heartbeat packets are set to maintain link connectivity. Data such as control schemes, traffic status, and equipment status are encrypted and transmitted. A unified data format and communication timing are agreed upon to ensure that data transmission latency meets the requirements of real-time traffic control.

[0048] In step S420, the traffic management system receives the resilience control scheme and generates a first traffic flow scheduling instruction and a second construction control map. Specifically, the traffic management system receives and parses the construction boundary adjustment sequence, lane control parameters, diversion ratio, intervention timing, and detour route information in the resilience control scheme, converting them into a first traffic flow scheduling instruction for roadside equipment, including variable message sign display content, traffic light phase, speed limit, and lane indicator status. Simultaneously, it generates a second construction control map containing the construction area location, closure range, warning distance, diversion route, and traffic prompts, providing a data foundation for traffic facility control and onboard terminal prompts.

[0049] Step S430: Traffic facility control is implemented according to the first traffic flow dispatching instruction, and the second construction control map is distributed to the vehicle platform port for real-time construction boundary prompts. Specifically, according to the timing requirements of the first traffic flow dispatching instruction, the traffic management system controls traffic facilities such as variable message signs, lane control signs, speed limit signs, and warning lights along the route to perform corresponding control actions. At the same time, the second construction control map is pushed to the vehicle navigation platform, roadside broadcasting unit, and mobile terminal application through the wireless communication network, providing drivers with real-time prompts on the construction boundary ahead, lane closure status, diversion routes, and speed limit requirements in a graphical and voice manner, thereby achieving dynamic traffic flow resilience control through vehicle-road cooperation.

[0050] This application employs a method of zoning and calibrating construction units and traffic sensing points within the construction area of ​​the traffic network to construct a road network construction topology. It then samples and predicts real-time traffic flow data using a traffic sensing network, obtains traffic congestion waves through wave feature vector transformation, and builds a resilience control unit at the road network edge based on the construction topology. This unit is composed of a first intervention component of the resilience capacity spectrum and a second equilibrium component of the zoning coupling matrix. The traffic congestion waves are then addressed through wave interference offsetting and detour diversion interventions. Through iterative allocation of the control task to each zone using a trust region framework, the control task is transformed into a resilience control scheme. The traffic management system responds to this scheme, enabling dynamic control and management of highway traffic flow resilience. This approach solves the technical problem of difficulty in real-time quantitative identification of congestion evolution patterns and precise, efficient intervention in dynamic control of highway traffic flow under complex construction scenarios. It achieves the technical effect of real-time quantitative identification of congestion evolution patterns and precise, efficient intervention of traffic flow.

[0051] In the above text, refer to Figure 1 This paper describes in detail a method for regulating the resilience of highway traffic flow under complex construction scenarios according to embodiments of the present invention. Next, reference will be made to... Figure 2 This invention describes a highway traffic flow resilience control system under complex construction scenarios according to embodiments of the present invention.

[0052] The highway traffic flow resilience control system for complex construction scenarios according to embodiments of the present invention addresses the technical problem of difficulty in real-time quantification and identification of congestion evolution patterns and implementation of precise and efficient intervention in existing dynamic control of highway traffic flow under complex construction scenarios. It achieves the technical effect of real-time quantification and identification of congestion evolution patterns and implementation of precise and efficient intervention in traffic flow. The highway traffic flow resilience control system for complex construction scenarios includes: a road network construction topology determination module 10, a traffic congestion wave determination module 20, a resilience control module 30, and a dynamic control management module 40.

[0053] The road network construction topology determination module 10 is used to perform zoning and traffic sensing point calibration of the construction area based on construction units in the traffic road network to determine the road network construction topology; the traffic congestion wave determination module 20 is used to perform sampling and prediction of real-time traffic flow data based on the traffic sensing network composed of traffic sensing points, and determine the traffic congestion wave by performing wave feature vector conversion; the resilience regulation module 30 is used to build a resilience regulation unit based on the road network construction topology, deploy it on the edge side of the road network, and solve the traffic congestion wave by intervention based on wave interference offsetting and detour diversion, and distribute it to each zone through iterative solution of the trust region framework, and transform to determine the resilience regulation scheme, wherein the resilience regulation unit is composed of a first intervention component based on the resilience capacity spectrum and a second equilibrium component based on the zoning coupling matrix; the dynamic regulation management module 40 is used to respond to the resilience regulation scheme through the traffic management system and perform dynamic regulation management of traffic flow resilience.

[0054] The specific configuration of the traffic congestion wave determination module 20 is described in detail below: As mentioned above, it performs sampling and prediction of real-time traffic flow data, and determines traffic congestion waves by performing wave feature vector conversion. The traffic congestion wave determination module 20 may further include: a traffic flow prediction unit for predicting real-time flow and direction of the real-time traffic flow data in a preset time zone based on the lightweight prediction unit, and determining traffic flow prediction data; a traffic evolution trajectory generation unit for generating traffic evolution trajectories with multiple traffic feature threads based on the real-time traffic flow data and traffic flow prediction data; and a wave feature vector conversion unit for performing wave feature vector conversion based on the traffic evolution trajectory to determine the traffic congestion wave.

[0055] The wave feature vector conversion unit may further include: a waveform element determination subunit for identifying the traffic evolution trajectory and determining waveform elements, wherein the amplitude is defined based on the difference between congestion density and free flow density, the wave speed is defined based on the speed at which the congestion wave peak propagates upstream, the wavelength is defined based on the product of the congestion duration and the wave speed, and the phase is defined based on the position of the congestion wave peak; and a traffic evolution trajectory reconstruction subunit is used to reconstruct the traffic evolution trajectory into the traffic congestion wave based on the waveform elements.

[0056] The specific configuration of the resilience control module 30 is described in detail below: As mentioned above, before building the resilience control unit, the resilience control module 30 may further include: a response surface construction unit for constructing a capacity-boundary response surface for each construction unit calibrated in the partition according to the engineering characteristics; and a resilience capacity spectrum construction unit for performing Gaussian process regression learning based on the response surface to construct a resilience capacity spectrum, wherein the output of the resilience capacity spectrum is the distribution of the passage capacity of the partition at different boundary positions.

[0057] The resilience control module 30, which is based on the road network construction topology, can further include: a partition coupling matrix establishment unit for establishing a partition coupling matrix based on the road network construction topology, wherein the diagonal elements are determined by the influence intensity of the change of the first partition boundary on the partition flow, and the off-diagonal elements are determined by the coupling transfer entropy of the influence of the change of the first partition boundary on the upstream and downstream adjacent partitions; and a component building unit for building a first intervention component based on the resilience capacity spectrum, building a second equilibrium component based on the partition coupling matrix, and cascading the first intervention component and the second equilibrium component to form the resilience control unit.

[0058] The resilience control module 30, which determines the resilience control scheme, may further include: a data initialization unit for receiving the traffic congestion wave and initializing the road network construction topology; an intervention solution unit for performing intervention solution based on wave interference offsetting and detour diversion according to the first intervention component to determine intervention wave parameters, wherein the intervention wave parameters include offsetting intervention waveform elements, diversion intervention waveform elements, and intervention timing; a trust domain framework iterative solution unit for performing trust domain framework iterative solution based on the intervention wave parameters under single-zone intervention allocation according to the second equalization component to determine zone intervention wave parameters; and a resilience control scheme determination unit for converting the zone intervention wave parameters into boundary adjustment sequences and detour parameters to determine the resilience control scheme.

[0059] After determining the resilience control scheme, the resilience control module 30 may further include: a digital control simulation unit for performing digital control simulation in the road network construction topology according to the resilience control scheme to determine control simulation data; a local intervention condition generation unit for identifying abnormal traffic flow characteristics in the control simulation data and generating local intervention conditions; and a scheme optimization unit for performing optimization solution based on the local intervention conditions to optimize the resilience control scheme.

[0060] The specific configuration of the dynamic control management module 40 is described in detail below: As mentioned above, the traffic management system responds to the resilience control scheme and performs dynamic control management of traffic flow resilience. The dynamic control management module 40 may further include: a communication interaction establishment unit for establishing communication interaction between the road network edge side and the traffic management system; a scheme receiving unit for the traffic management system to receive the resilience control scheme and generate a first traffic flow scheduling instruction and a second construction control map; and a traffic facility control unit for controlling traffic facilities according to the first traffic flow scheduling instruction and sending the second construction control map to the vehicle platform port for real-time construction boundary prompts.

[0061] After implementing dynamic regulation and management of traffic flow resilience, the resilience regulation module 30 may further include: a regulation effect evaluation unit for continuously monitoring traffic flow data in the traffic sensing network, performing differential comparison of traffic congestion waves before and after regulation, and evaluating the regulation effect of each zone, wherein the amplitude attenuation rate, wave velocity change and queue length reduction are used as quantitative evaluation standards; and an update learning unit for updating and learning the resilience regulation unit according to the regulation effect and the progress of zone engineering.

[0062] The highway traffic flow resilience control system under complex construction scenarios provided in this invention can execute the highway traffic flow resilience control method under complex construction scenarios provided in any embodiment of this invention, and has the corresponding functional modules and beneficial effects of the method.

[0063] Although this application makes various references to certain modules in the system according to the embodiments of this application, any number of different modules can be used and run on user terminals and / or servers. The various units and modules included are only divided according to functional logic, but are not limited to the above division, as long as the corresponding functions can be achieved; in addition, the specific names of each functional unit are only for easy distinction between each other and are not used to limit the scope of protection of this invention.

[0064] The above description is merely a preferred embodiment of the present invention and is not intended to limit the present invention in any way. Although the present invention has been disclosed above with reference to preferred embodiments, it is not intended to limit the present invention. Any person skilled in the art can make some modifications or alterations to the above-disclosed technical content to create equivalent embodiments without departing from the scope of the present invention. Any modifications, equivalent changes, and alterations made to the above embodiments based on the technical essence of the present invention without departing from the scope of the present invention shall still fall within the scope of the present invention.

Claims

1. A method for regulating the resilience of highway traffic flow under complex construction scenarios, characterized in that, The method includes: In the transportation network, the construction area is zoned and calibrated based on construction units and traffic sensing points to determine the construction topology of the road network; Based on the traffic sensing network formed by traffic sensing points, real-time traffic flow data is sampled and predicted, and traffic congestion waves are determined by wave feature vector conversion. Based on the road network construction topology, a resilience control unit is built and deployed on the edge of the road network. The traffic congestion wave is solved by intervention based on wave interference offsetting and detour diversion. The solution is distributed to each zone through iterative solution using a trust region framework. The resilience control scheme is then determined by transformation. The resilience control unit is composed of a first intervention component based on the resilience capacity spectrum and a second equalization component based on the zone coupling matrix. The traffic management system responds to the resilience control scheme by dynamically controlling and managing traffic flow resilience.

2. The method for regulating the resilience of highway traffic flow under complex construction scenarios as described in claim 1, characterized in that, Perform real-time traffic flow data sampling and prediction, and determine traffic congestion waves by performing wave eigenvector transformation, including: Based on the lightweight prediction unit, real-time traffic flow and direction prediction are performed on the real-time traffic flow data in a preset time zone to determine the traffic flow prediction data. Based on the real-time traffic flow data and traffic flow prediction data, a traffic evolution trajectory with multiple traffic feature threads is generated; Based on the traffic evolution trajectory, wave feature vector transformation is performed to determine the traffic congestion wave.

3. The method for regulating the resilience of highway traffic flow under complex construction scenarios as described in claim 2, characterized in that, Identify the traffic evolution trajectory and determine waveform elements, wherein the amplitude is defined based on the difference between congestion density and free flow density, the wave speed is defined based on the speed at which the congestion wave peak propagates upstream, the wavelength is defined based on the product of the congestion duration and the wave speed, and the phase is defined based on the position of the congestion wave peak. Based on waveform elements, the traffic evolution trajectory is reconstructed into the traffic congestion wave.

4. The method for regulating the resilience of highway traffic flow under complex construction scenarios as described in claim 1, characterized in that, Before constructing the resilience control unit, the following steps are included: For each construction unit categorized by zone, a capacity-boundary response surface is constructed based on the engineering characteristics; Gaussian process regression learning is performed based on the response surface to construct a resilience capacity spectrum, wherein the output of the resilience capacity spectrum is the distribution of the passage capacity of the partition at different boundary locations.

5. The method for regulating the resilience of highway traffic flow under complex construction scenarios as described in claim 4, characterized in that, Based on the aforementioned road network construction topology, a resilience control unit is constructed, including: Based on the road network construction topology, a partition coupling matrix is ​​established, wherein the diagonal elements are determined by the intensity of the impact of the change of the first partition boundary on the partition flow, and the off-diagonal elements are determined by the coupling transfer entropy of the impact of the change of the first partition boundary on the upstream and downstream adjacent partitions. A first intervention component is constructed based on the resilience capacity spectrum, and a second equalization component is constructed based on the partition coupling matrix. The first intervention component and the second equalization component are cascaded to form the resilience regulation unit.

6. The method for regulating the resilience of highway traffic flow under complex construction scenarios as described in claim 1, characterized in that, Determine resilience regulation strategies, including: The resilience control unit receives the traffic congestion wave and initializes the road network construction topology with data. Based on the first intervention component, an intervention solution based on wave interference offsetting and diversion is performed to determine the intervention wave parameters, wherein the intervention wave parameters include offsetting intervention waveform elements, diversion intervention waveform elements, and intervention timing; Based on the second equalization component, perform iterative solution of the trust region framework under single-partition intervention allocation based on the intervention wave parameters to determine the partition intervention wave parameters; The boundary adjustment sequence and bypass parameters of the partitioned intervention wave parameters are converted to determine the toughness control scheme.

7. The method for regulating the resilience of highway traffic flow under complex construction scenarios as described in claim 6, characterized in that, After determining the resilience regulation scheme, it includes: Based on the resilience control scheme, digital control simulation is performed in the road network construction topology to determine the control simulation data; Identify abnormal traffic flow characteristics in the control simulation data and generate local intervention conditions; Based on the local intervention conditions, an optimization solution based on the local intervention conditions is performed to optimize the resilience regulation scheme.

8. The method for regulating the resilience of highway traffic flow under complex construction scenarios as described in claim 1, characterized in that, The traffic management system responds to the aforementioned resilience control scheme by dynamically controlling and managing traffic flow resilience, including: Establish communication and interaction between the road network edge and the traffic management system; The traffic management system receives the resilience control scheme and generates a first traffic flow scheduling instruction and a second construction control map. Traffic facilities are controlled according to the first traffic flow scheduling instruction, and the second construction control map is sent to the vehicle platform port for real-time construction boundary prompts.

9. The method for regulating the resilience of highway traffic flow under complex construction scenarios as described in claim 1, characterized in that, After implementing dynamic regulation and management of traffic flow resilience, the following is included: The traffic sensing network continuously monitors traffic flow data, performs differential comparison of traffic congestion waves before and after regulation, and evaluates the regulation effect of each zone. Among them, the amplitude attenuation rate, wave speed change and queue length reduction are used as quantitative evaluation standards. Based on the aforementioned control effect and the progress of the zoning project, the resilience control unit is updated and learned.

10. A highway traffic flow resilience control system under complex construction scenarios, characterized in that, The system is used to implement the highway traffic flow resilience control method under complex construction scenarios as described in any one of claims 1-9, and the system includes: The road network construction topology determination module is used to perform zoning and traffic sensing point calibration of construction areas in the traffic road network based on construction units, and to determine the road network construction topology. The traffic congestion wave determination module is used to perform sampling and prediction of real-time traffic flow data based on the traffic sensing network composed of traffic sensing points, and to determine the traffic congestion wave by performing wave feature vector conversion. The resilience control module is used to build a resilience control unit based on the road network construction topology and deploy it on the edge of the road network. It solves the traffic congestion wave by intervening and diverting based on wave interference and diversion, and distributes it to each partition through iterative solution using the trust region framework. The transformation determines the resilience control scheme. The resilience control unit is composed of a first intervention component based on the resilience capacity spectrum and a second equalization component based on the partition coupling matrix. The dynamic control and management module is used to respond to the resilience control scheme through the traffic management system and perform dynamic control and management of traffic flow resilience.

Images