Highway dynamic regional variable speed limit method and system based on hybrid model

Abstract

The application provides a highway dynamic region variable speed limit method and system based on a hybrid model, and belongs to the technical field of traffic control; the method comprises the following steps: determining congestion cells by preprocessing traffic information in a target road section of a highway; setting a current variable speed limit control region and recording the length of the region, adjusting the position of the congestion cells by comparing the length of the region, and calculating the final length of each sub-region in the variable speed limit control region; constructing a cell transmission model and extending the model; identifying the nonlinear characteristics of the extended cell transmission model and processing the characteristics into a linear constraint set to construct a hybrid model; setting a variable speed limit control target, converting the control problem into a mixed integer linear programming problem, and issuing the obtained optimal speed limit value to vehicles in the target road section; and the application can significantly improve the space-time adaptability and dynamic response capability of the highway dynamic region variable speed limit control.

Description

TECHNICAL FIELD

[0001] The present application belongs to the technical field of traffic control, and particularly relates to a method and system for dynamically variable speed limit of expressway based on a hybrid model. BACKGROUND

[0002] The statements in this section merely provide background information related to the present application and do not necessarily constitute the prior art.

[0003] Variable speed limit control, as one of the key means of active traffic management on expressways, mainly adjusts the speed limit value to relieve traffic congestion and improve road traffic efficiency and safety. Traditional VSL systems usually rely on roadside variable information signs to issue speed limit instructions. Although this method can adjust traffic flow to some extent, it still has obvious limitations: first, it requires large-scale deployment of physical devices, which is costly and has limited coverage, especially in weak facilities or complex terrain sections; second, its control area is usually static and preset, and cannot be dynamically adjusted according to the real-time spatial evolution of traffic congestion, resulting in insufficient ability to deal with sudden congestion or congestion migration.

[0004] With the development of Internet of Vehicles, wireless communication and intelligent sensing technology, a control method has emerged that directly issues dynamic speed limit information to drivers through vehicle terminals or mobile navigation applications. This method to some extent breaks away from the dependence on fixed facilities and improves the flexibility and real-time performance of information dissemination. In this context, dynamically delineating speed limit areas based on predicted traffic states and combining macro traffic flow models for control decisions has become an important direction for the development of VSL systems. Among them, the cell transmission model is often combined with model predictive control strategies to form a real-time control framework for expressways with strong interpretability and predictive control capabilities due to its effective description of the dynamic characteristics of traffic flow.

[0005] However, dynamic area variable speed limit control puts higher requirements on models and algorithms. Discrete events in the model, such as discrete speed limit values, selection and division of speed limit areas, piecewise linear relationships of flow and density, etc., belong to a hybrid characteristic, causing high nonlinearity of the control problem, and thus existing model predictive control methods are difficult to solve and achieve dynamic response. SUMMARY

[0006] To overcome the shortcomings of the prior art, the present application provides a method and system for dynamically variable speed limit of expressway based on a hybrid model, which can significantly improve the spatiotemporal adaptability and dynamic response capability of the variable speed limit control of the dynamic area of the expressway.

[0007] To achieve the above purpose, one or more embodiments of the present application provide the following technical solutions:

[0008] The application provides a method for dynamically setting variable speed limits in a highway region based on a hybrid model.

[0009] The method for dynamically setting variable speed limits in a highway region based on a hybrid model comprises the following steps.

[0010] Traffic information in a target section of a highway is obtained, and congested cells in the target section are determined by preprocessing the obtained traffic information.

[0011] A current variable speed limit control region is set, and the length of the region is recorded. The position of the congested cells is adjusted according to the length of the region, and the final length of each subregion in the variable speed limit control region is calculated.

[0012] A cell transmission model is constructed, and the constructed cell transmission model is extended based on the flow conservation theory according to the final length of the obtained subregion.

[0013] Nonlinear characteristics of the extended cell transmission model are identified, and linearization processing is performed to obtain a linear constraint set, so as to construct a hybrid model. A variable speed limit control target is set based on the constructed hybrid model. The control problem of the variable speed limit control target is converted into a mixed integer linear programming problem, and the obtained optimal speed limit value is sent to vehicles in the target section.

[0014] Further, the preprocessing comprises the following steps: the target section is divided into a plurality of initial subregions in units of a preset minimum cell length.

[0015] Further, the adjustment of the position of the congested cells and the calculation of the final length of each subregion in the variable speed limit control region comprise the following steps: the length of the control region is determined by extremum comparison, and the obtained length of the control region is taken as the region length to calculate the number of initial subregions. The position of the cell boundary is adjusted according to the ratio of the obtained region length and the number of initial subregions, and the final length of each subregion is obtained.

[0016] Further, the extension of the constructed cell transmission model comprises the following steps: the final length of the obtained subregion is introduced into the cell transmission model together with a variable speed limit value and a speed limit compliance rate to extend the cell transmission model.

[0017] Further, the linearization processing comprises the following steps: different linearization methods are adopted for different types of nonlinear characteristics, and the nonlinear characteristics of the extended cell transmission model are converted into a linear constraint set.

[0018] Further, the variable speed limit control target comprises a first optimization target and a second optimization target; wherein the first optimization target is to minimize the sum of differences between each cell density and critical density, and the second optimization target is to minimize the time variation degree of the recommended speed limit value.

[0019] Further, the step of issuing the optimal speed limit value to the vehicles in the target road section comprises: solving the mixed integer linear programming problem by using a solver to obtain the optimal control sequence in the future control time domain; and issuing the first control quantity in the optimal control sequence as the optimal speed limit value to the vehicles in the target road section.

[0020] The second aspect of the present application provides a highway dynamic regional variable speed limit system based on a hybrid model.

[0021] The highway dynamic regional variable speed limit system based on the hybrid model comprises:

[0022] The preprocessing module is configured to: obtain traffic information in a target road section of the highway, and determine congestion cells in the target road section of the highway by preprocessing the obtained traffic information.

[0023] The speed limit region selection and cell division module is configured to: set a current variable speed limit control region, and record the length of the region; adjust the positions of the congestion cells and calculate the final length of each sub-region in the variable speed limit control region by comparing the lengths of the regions.

[0024] The cell transmission model expansion module is configured to: construct a cell transmission model, and expand the constructed cell transmission model based on the flow conservation theory according to the final lengths of the sub-regions.

[0025] The hybrid model construction module is configured to: identify the nonlinear characteristics of the expanded cell transmission model, and linearize the nonlinear characteristics into a linear constraint set for constructing a hybrid model.

[0026] The variable speed limit control module is configured to: set a variable speed limit control target based on the constructed hybrid model.

[0027] The optimal speed limit value solving module is configured to: convert the control problem of the variable speed limit control target into a mixed integer linear programming problem, and issue the obtained optimal speed limit value to the vehicles in the target road section.

[0028] The third aspect of the present application provides a computer readable storage medium having a program stored thereon, wherein the program is executed by a processor to implement the steps in the highway dynamic regional variable speed limit method based on the hybrid model according to the first aspect of the present application.

[0029] The fourth aspect of the present application provides an electronic device, comprising a memory, a processor and a program stored in the memory and executable on the processor, wherein the processor implements the steps in the method for dynamically variable speed limit of expressway based on a hybrid model according to the first aspect of the present application when executing the program.

[0030] The above one or more technical solutions have the following beneficial effects:

[0031] The present application constructs a cell transmission model, and extends the constructed cell transmission model according to the final length of the obtained sub-area based on the flow conservation theory; the nonlinear characteristics of the extended cell transmission model are identified and linearized into a linear constraint set, so as to construct a hybrid model. The original highly nonlinear control model is converted into a mixed integer linear programming problem, which effectively overcomes the bottleneck that the existing model predictive control is difficult to solve when processing such systems, greatly improves the solvability and practicability of the model, and can significantly improve the spatio-temporal adaptability and dynamic response capability of the dynamic regional variable speed limit control of the expressway.

[0032] Advantages of the additional aspects of the present application will be partially given in the following description, partially will become obvious from the following description, or will be known by the practice of the present application. BRIEF DESCRIPTION OF DRAWINGS

[0033] The accompanying drawings, which form a part of the present application, are used to provide further understanding of the present application, and the illustrative embodiments of the present application and their description serve the purpose of explaining the present application, and do not constitute improper limitations on the present application.

[0034] Figure 1 The flow chart of the method for dynamically variable speed limit of expressway based on a hybrid model in the embodiment one of the present application. DETAILED DESCRIPTION

[0035] It should be noted that the following detailed description is exemplary, and is intended to provide further explanation of the present application. Unless otherwise specified, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which the present application belongs.

[0036] It should be noted that the terms used herein are only for the purpose of describing specific embodiments, and are not intended to limit the exemplary embodiments according to the present application.

[0037] In the case of no conflict, the embodiments in the present application and the features in the embodiments can be combined with each other.

[0038] In order to facilitate the understanding of the technical solutions of the present application, the following terms are explained:

[0039] Variable Speed Limit (VSL), a traffic management strategy that optimizes traffic flow by dynamically adjusting speed limits on roads to reduce congestion and improve safety. Speed limits can be adjusted based on real-time traffic conditions, weather conditions, or road construction, among other factors, to achieve more efficient traffic management.

[0040] Cell Transmission Model (CTM), a mathematical model used to describe the dynamic behavior of traffic flow. The model divides roads into discrete cells and assumes that traffic flows between these cells according to certain rules. By simulating the transmission process between cells, CTM can predict parameters such as traffic density, speed, and flow, and is widely used in traffic flow theory and traffic engineering.

[0041] Model Predictive Control (MPC), an advanced control strategy that predicts the behavior of a system over a certain period of time based on the system model, and optimizes control inputs to achieve the desired control objective. MPC solves an optimization problem within a finite time range to determine the best control action at the current time, taking into account the dynamic characteristics and constraints of the system. It has wide applications in industrial process control, traffic system management and other fields.

[0042] Mixed-Integer Linear Programming (MILP), a mathematical optimization method used to solve linear programming problems containing continuous variables and integer variables. MILP models describe problems through linear objective functions and linear constraints, where integer variables represent discrete choices in decision-making processes, such as switch states, path selection, etc. MILP has important applications in traffic management, scheduling optimization, logistics planning and other fields, and can effectively solve complex optimization problems.

[0043] The overall idea provided by the application is: the application provides a highway dynamic region variable speed limit method based on a hybrid model, which first generates and adjusts a speed limit control region dynamically based on predicted congestion information through an adaptive algorithm, so as to better match the spatial position and range of congestion; then, a cell transmission model is expanded, that is, factors such as a variable speed limit value, a variable cell length, and a speed limit compliance rate are introduced into the model, and the model is built into a hybrid model; finally, based on the predicted congestion information, the hyperparameters (such as the number of cells, the length of cells, the range of speed limit values, and control targets) of the variable speed limit control model are set; and based on the hybrid model prediction control method, the control problem of each step is converted into a mixed integer linear programming problem. Therefore, the application can balance the calculation speed while ensuring the control accuracy through reasonable parameter adjustment, and has the advantages of strong foresight, strong interpretability, and strong expandability.

[0044] Embodiment one

[0045] The embodiment discloses a highway dynamic region variable speed limit method based on a hybrid model.

[0046] As shown in Figure 1 The highway dynamic region variable speed limit method based on a hybrid model comprises the following steps:

[0047] In step S1, traffic information in a target road section of a highway is acquired, and congestion cells in the target road section of the highway are determined by preprocessing the obtained traffic information.

[0048] In step S2, a current variable speed limit control region is set, and the length of the region is recorded; the position of the congestion cells is adjusted through the length comparison, and the final length of each subregion in the variable speed limit control region is calculated.

[0049] In step S3, a cell transmission model is constructed, and the constructed cell transmission model is expanded according to the final length of the obtained subregion based on the flow conservation theory.

[0050] In step S4, the nonlinear characteristics of the expanded cell transmission model are identified, and linearization processing is performed to obtain a linear constraint set, so as to construct a hybrid model; a variable speed limit control target is set based on the constructed hybrid model; the control problem of the variable speed limit control target is converted into a mixed integer linear programming problem, and the obtained optimal speed limit value is sent to vehicles in the target road section.

[0051] Based on the above process, the application can significantly improve the spatiotemporal adaptability and dynamic response capability of the highway dynamic region variable speed limit control. In order to facilitate the understanding of the technical scheme of the application, the specific implementation method in the technical scheme of the application is further explained and described below.

[0052] In step S1, traffic information in the target section of the expressway is acquired, and congested cells in the target section of the expressway are determined by preprocessing the obtained traffic information.

[0053] Firstly, traffic information containing various fixed information (representing cell transmission model parameters) in the target section of the expressway is imported, i.e., section topology information, macroscopic traffic flow parameters, discrete speed limit value set, and real-time information (representing control model external input, imported in the form of a vector); wherein the real-time information includes traffic demand, downstream boundary condition, traffic flow state, and the like.

[0054] In the given section range, the section is divided into initial sub-regions in the minimum cell unit; the number of sub-regions should be less than or equal to the number of cells, i.e., a sub-region can be a cell or can be divided into multiple cells, and in this embodiment, the number of the two is equal by default. Wherein the length of the minimum cell unit is represented as .

[0055] Subsequently, the basic cell transmission model is used to predict traffic density in the future steps, and all cells are determined as congested cells. Wherein represents the critical density of cell , represents the control time domain number, represents the traffic density of cell .

[0056] In step S2, the current variable speed control region is set, and the region length is recorded; the variable speed control region is determined according to the congestion range predicted in step S1, and the final length of each sub-region is calculated.

[0057] The most upstream congested cell is set as the starting point, and the most downstream congested cell is set as the ending point, the current variable speed control region is set, and the region length is recorded as ; and should be an integer multiple of and rounded up.

[0058] Firstly, it is judged whether the control region length exceeds , if so, ; it is judged whether the control region is less than , if so, . Wherein represents the maximum length of the control region, represents the minimum length of the control region.

[0059] Subsequently, the number of initial sub-regions is calculated:​ .thus, Initialized to the maximum allowed value, theoretically, the finer the cell division, the higher the model accuracy, but it also increases the complexity of real-time control problems. Therefore, it is necessary to limit the maximum number of cells to avoid excessively slow real-time control response. Specifically: [The text abruptly ends here, likely due to an incomplete sentence or missing information.] Is it true? If it is true, then... ;in, This indicates the maximum number of cells.

[0060] Thus, the initial length of each sub-region can be obtained. .in, Indicates the first The initial length of each sub-region. The position of the cell boundary is adjusted according to factors such as the fixed speed limit signs and terrain of the road section. That is, when the fixed speed limit signs and special terrain (such as the junction of the merging and diverging areas, sharp bends and steep slopes) are divided inside the cell, the cell should be divided so that it is located at the cell boundary.

[0061] Finally, the final length of each sub-region is obtained. ,in, Indicates the first The final length of each sub-region. And, every... For each time step, the speed limit zone range is updated once, that is, step S2 is repeated.

[0062] In step S3, a cell transmission model is constructed, and the constructed cell transmission model is extended based on the final length of the obtained sub-region according to the flow conservation theory.

[0063] The flow conservation theory can be expressed as:

[0064] (1)

[0065] in, For discrete time steps, Indicates a time step. express Time Cell Flow to cells Traffic flow. When using the above formula, for any cell It must satisfy the Courant-Friedrichs-Lewy (CFL) condition, that is... Otherwise, the formula would fail to satisfy both numerical stability and physical rationality in simulating traffic flow. Therefore, and It should be set up reasonably; among them, Represents cell The free flow velocity.

[0066] Further, on the main line of the highway The calculation formula is:

[0067] (2)

[0068] Wherein, is the maximum traffic flow that can be sent by the cell in the free flow state; is the maximum traffic flow that can be received by the cell , that is,

[0069] (3)

[0070] (4)

[0071] Wherein, represents the jam density of the cell , at which the number of vehicles accommodated by the cell reaches the limit, and all vehicles cannot move; represents the congestion wave propagation speed of the cell ; represents the actual speed limit value at time t, and the relationship between the actual speed limit value and the recommended speed limit value is:

[0072] ;

[0073] Wherein, is the compliance factor, indicating that the actual speed limit value of the traffic flow may be lower or higher than the recommended speed limit value given by the control model; The value of is dynamically calculated by the control model according to real-time traffic conditions. Specifically, ; wherein S represents a set of discrete speed limit values; is the th (ascending order) available speed limit value. Note that the upper bound of is the free flow speed .

[0074] The calculation formula is:

[0075] (5)

[0076] The calculation formula is:

[0077] (6)

[0078] Generally, the above formula is only applicable to ​​the case. To consider the boundary conditions at the upstream and downstream, one can extend the model to ; in this case, the virtual cells 0 and 1 represent the upstream and downstream boundaries of the link, respectively.

[0079] In step S4, the nonlinear features of the extended cell transmission model are identified and linearized into a set of linear constraints to construct a hybrid model; based on the constructed hybrid model, a variable speed limit control objective is set; the control problem of the variable speed limit control objective is converted into a mixed integer linear programming problem, and the optimal speed limit value obtained is issued to the vehicles in the target link. Specifically, the following methods can be used to achieve this:

[0080] Step S4-1, identify the nonlinear features of the extended cell transmission model, and linearize it into a set of linear constraints to construct a hybrid model.

[0081] In step S3, a cell transmission model for variable speed limit control is constructed, which has the following several sources of nonlinear features:

[0082] 1) the bilinear term in equation (3) can be linearized by McCormick envelope method.

[0083] 2) the nonlinear nature of the functions , , in equations (2), (3), and (4) is a piecewise affine dynamic driven by discrete events, which can be converted into linear constraints by Big-M method.

[0084] 3) the nonlinear features in equations (5) and (6) are variants of the functions and , respectively, which can be approximated by piecewise linearization; where denotes the independent variable of the function.

[0085] 4) the decision variable belongs to a discrete set, which can be linearized by introducing an indicator variable.

[0086] Further, the linearization method of the bilinear term in equation (3) is as follows (the time step k is omitted):

[0087] (7)

[0088] (8)

[0089] (9)

[0090] (10) ​​

[0091] where, is a convex relaxation term with upper bound and representing the upper bound and lower bound, respectively. The theoretical lower bound of and is 0, and the upper bound is and respectively. Therefore, formula (3) can be updated as:

[0092] (11)

[0093] The error of McCormick envelope method can be reduced by tightening the upper and lower bounds. In addition, a more commonly used method is to use piecewise McCormick envelope method, which can further improve the accuracy of linearization approximation by dividing the value range of variables into multiple intervals and applying McCormick envelope in each interval.

[0094] Further, the linearization method of formula (2) is as follows:

[0095] Introduce auxiliary variables , and ; where, and are binary variables, is a large enough constant. Using the Big-M method, linearization is achieved through the following constraints:

[0096] (12)

[0097] (13)

[0098] (14)

[0099] (15)

[0100] (16)

[0101] (17)

[0102] In the above formulas, and are indicator variables of and respectively.

[0103] Further, the linearization method of formula (3) is as follows:

[0104] Introduce auxiliary variables , and ; wherein, and are binary variables, is a sufficiently large constant. Using Big-M method, linearize by the following constraints:

[0105] (18)

[0106] (19)

[0107] (20)

[0108] (21)

[0109] (22)

[0110] (23)

[0111] (24)

[0112] (25)

[0113] In the above equations, and are the indicator variables of and respectively.

[0114] It can be understood that the linearization method of equation (4) is the same as equation (3).

[0115] Further, the nonlinear characteristics in equation (5) (6) are the variants of functions and respectively, which can be approximated by simple piecewise linearization. In this model, the actual lower bound of the speed limit value given by the control model will not generally exceed , and thanks to this range of values, even if a straight line is used for approximation, the error is still at a low level. For example, substituting , into equation (5) obtains two points ), ( ), and the line connecting the two points is a single piecewise linear fitting.

[0116] It can be understood that the linearization method of equation (6) is the same. By increasing the number of segments, the fitting accuracy can be significantly improved, but at the same time, new variables and constraints are introduced, increasing the complexity of the model. The number of segments of the piecewise linearization method needs to be considered in the balance of accuracy and computational efficiency.

[0117] For the set , introduce an indicator variable , . Where denotes the number of selectable speed limit values; , is a logical variable, 0 means the speed limit value is not selected, 1 means the speed limit value of the cell is , i.e.:

[0118] (26)

[0119] This indication relationship can be realized by the following linearization constraints:

[0120] (27)

[0121] (28)

[0122] So far, all the nonlinear features in the above constraints have been linearized.

[0123] Step S4-2, set the variable speed limit control target based on the built hybrid model.

[0124] Let the vector represent the density of the cell, the vector represent the critical density of the cell; Let the vector represent the recommended speed limit value of the cell. Let represent the control time domain (the default prediction time domain number is the same as the control time domain); Based on model predictive control, the variable speed limit control problem at time can be represented as the following optimization problem:

[0125] (29)

[0126] All related equality constraints of the cell transmission model in step S3 and all equality and inequality constraints for linearization in step four:

[0127]

[0128] (30)

[0129] (31)

[0130] (32)

[0131] (33)

[0132] (34)

[0133] (35)

[0134] Equation (29) represents the control objective of variable speed limit, which has two terms. The first term is to minimize the sum of the difference between the density of each cell and its critical density. This is a common objective in macroscopic traffic flow control, aiming to make the density of each cell as close to its critical density as possible, thereby maximizing the capacity of the road and reducing congestion. The second term is to minimize the degree of temporal variation of the recommended speed limit value. This objective aims to smooth the traffic flow, reduce the confusion and discomfort of drivers caused by frequent or large changes in speed limit, and thus improve the safety and comfort of driving. Frequent changes in speed limit can increase the risk of accidents, so when designing a variable speed limit control system, a balance needs to be found between optimizing traffic flow and ensuring driving safety. Further, 、 and are weight coefficients; and are used to adjust the weight relationship between the two optimization objectives, aiming to avoid instability or oscillation that may occur when the density of the cell approaches the critical density.

[0135] Equations (30), (31) specify the upper and lower bounds of and , denotes the density lower bound of the cell , denotes the density upper bound of the cell , denotes the speed limit value lower bound, denotes the speed limit value upper bound. and both have a theoretical lower bound of 0, and an upper bound of 、 .

[0136] Equations (32), (33) limit the degree of variation of the speed limit value in time and space, used to ensure the stability of the traffic flow; denotes the maximum allowed speed limit variation value of any cell in two adjacent time steps, denotes the maximum allowed speed difference value between two adjacent cells at any time.

[0137] Equations (34), (35) specify the upstream and downstream boundary conditions, denotes the density of the upstream boundary cell, denotes the density of the downstream boundary cell. Density value of the upstream virtual cell corresponding to the system external input. Density of the downstream boundary virtual cell corresponding to the system external input, which is calculated by the predicted traffic demand.

[0138] Step S4-3, convert the control problem of the variable speed limit control target into a mixed integer linear programming problem, and issue the optimal speed limit value obtained to the vehicle in the target road section.

[0139] After completing the linearization process, the controllable problem proposed in the above method has been successfully converted into a MILP problem. Such problems can be solved by applying existing mature solving techniques, including commercial solvers (such as GRUOBI and CPLEX), and open source solvers (such as CBC). These solvers can efficiently handle complex optimization problems and provide global optimal solutions. In each time step , the optimization problem of formula (29) is solved, and the optimal control sequence is , only the first control input is applied to the actual system, so the MPC is a receding horizon optimization method, which can significantly reduce the impact of prediction model inaccuracy and traffic demand changes.

[0140] In addition, the present application realizes that, since the speed limit information needs to be updated every few minutes, the solving efficiency is one of the key factors of variable speed limit control. Although existing solvers are quite mature, when the number of cells and the control time domain increase, the complexity of the MILP problem will increase exponentially, and when dealing with large-scale variable speed limit control problems, it may face limitations in computing resources and time. Therefore, one of the future directions of technological development is to use domain knowledge to design specialized acceleration algorithms to reduce the size of MILP to improve its solving speed, such as customized heuristic methods, data-driven machine learning methods, etc. Through these methods, the solving time can be significantly reduced, and the real-time performance and response speed of the system can be improved, so that the variable speed limit control strategy of the present application is more practical and competitive in a wider range of application scenarios.

[0141] Based on the above method, the application proposes an innovative hybrid model predictive control method for the application scenario of dynamic regional variable speed limit control on expressways. This method is suitable for the following scenarios: using perception and prediction technology to identify traffic congestion key points and dynamically setting speed limit areas upstream of them. By combining macroscopic traffic flow models with advanced control algorithms, the application can calculate the optimal speed limit value and transmit the speed limit information to vehicles in real time through communication means such as vehicle networking and mobile navigation. This method effectively solves the problems of fixed speed limit areas and insufficient coverage in traditional variable speed limit control, significantly improving the spatiotemporal adaptability and dynamic response capability of variable speed limit control.

[0142] The hybrid model predictive control framework proposed in the application not only maintains the integrity of macroscopic traffic flow dynamics, but also cleverly uses its hybrid characteristics to convert complex control problems into mixed integer linear programming problems. This transformation enables the problem to obtain a globally optimal solution with existing mature solvers or algorithms, thereby improving the solution efficiency and control accuracy. In addition, the framework has high interpretability and can be extended to coordinated control with other real-time traffic control to achieve more comprehensive traffic management strategies.

[0143] In summary, by introducing the hybrid model predictive control method, the application not only improves the efficiency and effectiveness of dynamic regional variable speed limit control on expressways, but also provides a technical foundation and innovative idea for the further development of future intelligent transportation systems.

[0144] Embodiment Two

[0145] The embodiment discloses a hybrid model-based dynamic regional variable speed limit system for expressways.

[0146] The hybrid model-based dynamic regional variable speed limit system for expressways comprises:

[0147] The preprocessing module is configured to obtain traffic information within a target road section of an expressway, and determine congestion cells within the target road section of the expressway by preprocessing the obtained traffic information.

[0148] The speed limit area selection and cell division module is configured to set a current variable speed limit control area and record the length of the area, adjust the position of the congestion cells based on the length of the area, and calculate the final length of each sub-area in the variable speed limit control area.

[0149] The cell transmission model expansion module is configured to construct a cell transmission model and expand the constructed cell transmission model based on the final length of the obtained sub-area according to the flow conservation theory.

[0150] The hybrid model construction module is configured to identify the nonlinear characteristics of the extended cellular transmission model and linearize the nonlinear characteristics into a set of linear constraints to construct a hybrid model.

[0151] The variable speed limit control module is configured to set a variable speed limit control target based on the constructed hybrid model.

[0152] The optimal speed limit value solving module is configured to convert the control problem of the variable speed limit control target into a mixed integer linear programming problem and issue the obtained optimal speed limit value to vehicles in the target road section.

[0153] Embodiment three

[0154] An object of the embodiment is to provide a computer-readable storage medium.

[0155] A computer-readable storage medium having stored thereon a computer program, which, when executed by a processor, implements the steps in the method for dynamically setting variable speed limits in a highway region based on a hybrid model according to Embodiment One of the present disclosure.

[0156] Embodiment four

[0157] An object of the embodiment is to provide an electronic device.

[0158] An electronic device includes a memory, a processor, and a program stored in the memory and executable on the processor, and the processor implements the steps in the method for dynamically setting variable speed limits in a highway region based on a hybrid model according to Embodiment One of the present disclosure when executing the program.

[0159] The steps and methods involved in the above embodiments two, three, and four correspond to Embodiment One, and the specific implementation can be referred to the relevant description in Embodiment One. The term "computer-readable storage medium" should be understood to include a single medium or multiple media of one or more instruction sets; it should also be understood to include any medium that can store, encode, or carry instruction sets for execution by a processor and cause the processor to perform any of the methods in the present disclosure.

[0160] Those skilled in the art should understand that each module or step of the present disclosure described above can be implemented by a general-purpose computer device, and alternatively, they can be implemented by program codes executable by a computing device, so that they can be stored in a storage device for execution by a computing device, or they can be respectively manufactured into individual integrated circuit modules, or a plurality of modules or steps among them can be manufactured into a single integrated circuit module. The present disclosure is not limited to any specific combination of hardware and software.

[0161] The above describes the specific embodiments of the present application in combination with the drawings, but is not a limitation on the protection scope of the present application. Those skilled in the art should understand that various modifications or variations made by those skilled in the art on the basis of the technical solutions of the present application without creative labor are still within the protection scope of the present application.

Claims

1. A method for dynamic regional variable speed limits on highways based on a hybrid model, characterized in that, include: Obtain traffic information within the target section of the highway, and determine the congestion cells within the target section of the highway through preprocessing of the obtained traffic information; Set the current variable speed limit control zone and record the zone length; adjust the position of congestion cells by comparing the zone lengths and calculate the final length of each sub-zone in the variable speed limit control zone; A cellular transmission model is constructed, and based on the flow conservation theory, the constructed cellular transmission model is extended according to the final length of the obtained sub-regions; The nonlinear characteristics of the extended cellular transport model are identified and linearized into a set of linear constraints to construct a hybrid model. The linearization process involves using corresponding linearization methods for different types of nonlinear characteristics to transform them into a set of linear constraints. Specifically, the McCormick envelope method is used for nonlinear characteristics with bilinear terms; piecewise affine dynamics driven by discrete events are converted into linear constraints using the Big-M method; and nonlinear characteristics that are functions are further simplified. and The variant is approximated by piecewise linearization; Based on the established hybrid model, a variable speed limit control objective is set. Specifically, let the vector... Represents the density of a cell, a vector Let the vector represent the critical density of the cell; Indicates the suggested speed limit value for a cell; let This represents the control time domain, which has the same number of control time domains as the default prediction time domain; based on model predictive control, in The variable speed limiting control problem at any given time can be expressed as the following optimization problem: ； in, , and Indicates the weighting coefficient; Represents the first in the optimal control sequence One control input; The equality constraints under the cellular transmission model and the linearized equality and inequality constraints under the set variable speed limit control objective are expressed as follows: ； ； ； ； ； ； in, This represents the lower bound of the constraint. Indicates the upper bound of the constraint; Represents any cell The maximum permissible change in speed limit between two adjacent moments. This represents the maximum permissible velocity difference between two adjacent cells at any given time. This represents the density of the upstream boundary cell. This represents the density value of the upstream virtual cell; This represents the density of the downstream boundary cell. This represents the density of the downstream boundary virtual cells; The control problem of variable speed limit control is transformed into a mixed integer linear programming problem, and the obtained optimal speed limit value is issued to vehicles in the target road segment; the variable speed limit control objective includes a first optimization objective and a second optimization objective; wherein, the first optimization objective is to minimize the sum of the differences between each cell density and the critical density, and the second optimization objective is to minimize the time variation of the suggested speed limit value.

2. The method for dynamic regional variable speed limit on highways based on a hybrid model as described in claim 1, characterized in that, The preprocessing includes: dividing the target road segment into multiple initial sub-regions using a preset minimum cell length as the unit; then, predicting the traffic density of the target road segment within multiple preset steps in the future, and determining whether it is a congested cell by comparing the obtained traffic density with a preset critical density.

3. The method for dynamic regional variable speed limit on highways based on a hybrid model as described in claim 1, characterized in that, The adjustment of congestion cell positions and the calculation of the final length of each sub-region in the variable speed limit control area include: determining the control area length by comparing extreme values, and using the obtained control area length as the area length to calculate the initial number of sub-regions; adjusting the cell boundary position according to the ratio of the obtained area length to the initial number of sub-regions and obtaining the final length of each sub-region.

4. The method for dynamic regional variable speed limits on highways based on a hybrid model as described in claim 1, characterized in that, The established cellular transport model is extended by incorporating the final length of the obtained sub-region, along with the variable rate limit value and rate limit compliance rate, into the cellular transport model.

5. The method for dynamic regional variable speed limits on highways based on a hybrid model as described in claim 1, characterized in that, Sending the optimal speed limit to vehicles in the target road segment includes: solving a mixed-integer linear programming problem using a solver to obtain the optimal control sequence in the future control time domain; and sending the first control variable in the optimal control sequence as the optimal speed limit to vehicles in the target road segment.

6. A dynamic regional variable speed limit system for highways based on a hybrid model, employing the variable speed limit method as described in any one of claims 1-5, characterized in that, include: The preprocessing module is configured to: acquire traffic information within the target section of the highway, and determine congestion cells within the target section of the highway by preprocessing the acquired traffic information; The speed limit zone selection and cell division module is configured to: set the current variable speed limit control zone and record the zone length; adjust the position of congestion cells by comparing the zone lengths and calculate the final length of each sub-zone in the variable speed limit control zone; The cell transmission model extension module is configured to: construct a cell transmission model and extend the constructed cell transmission model based on the final length of the obtained sub-region according to the flow conservation theory; The hybrid model building module is configured to: identify the nonlinear characteristics of the extended cell transport model and linearize them into a set of linear constraints to build the hybrid model; The variable speed limit control module is configured to set a variable speed limit control target based on the established hybrid model; The optimal speed limit value solution module is configured to: transform the control problem of variable speed limit control target into a mixed integer linear programming problem, and distribute the obtained optimal speed limit value to vehicles in the target road segment.

7. A computer-readable storage medium having a program stored thereon, characterized in that, When executed by the processor, the program implements the steps in the hybrid model-based dynamic regional variable speed limit method for highways as described in any one of claims 1-5.

8. An electronic device, comprising a memory, a processor, and a program stored in the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the hybrid model-based dynamic regional variable speed limit method for highways as described in any one of claims 1-5.

Images