A traffic status control method and system for highway toll station areas

Abstract

The present invention belongs to the technical field of traffic management, and provides a method and system for controlling the traffic state in the toll station area of a highway. The method includes: obtaining traffic data of the toll station and the upstream gantry; based on the traffic data of the toll station and the upstream gantry, using an LSTM neural network model to predict the traffic demand of the toll station for a period of time; based on the traffic demand, using a cell transmission model to obtain the traffic flow operation state of the toll station for a period of time, where the traffic flow operation state includes traffic volume and traffic density; based on the traffic flow operation state and the number of lanes, constructing an objective function for the total driving time of vehicles in the toll station area, and combining with constraint conditions, using model predictive control to perform rolling and iterative optimization calculations to obtain the speed change limit and lane configuration corresponding to the minimum total driving time. The present invention can significantly improve the traffic performance of the toll station and improve the traffic efficiency of the highway.

Description

Technical Field

[0001] The present invention belongs to the technical field of traffic management, and in particular relates to a method and system for controlling traffic status in a highway toll station area. Background Art

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

[0003] Existing research mainly starts from the perspective of station design and optimizes the number of lane openings and closings based on the operating costs of toll plazas. Compared with urban traffic, although the traffic flow at highway toll stations has certain regularity, it is more random, and the proportion of ETC and MTC vehicles is also time-varying. Static lane configuration cannot achieve real-time opening and closing control and effective use of road resources, and the application scenarios are limited.

[0004] Yuan et al. proposed a VSL-RM integrated control strategy to improve the efficiency of highway trunk lines. The strategy coordinates the traffic from trunk lines and toll booths and adjusts the traffic density in the trunk line merging area. However, the purpose of this approach is to improve the traffic efficiency of trunk lines, not toll booths. Therefore, active traffic control at highway toll booths needs further study: static lane configuration cannot dynamically adapt to real-time traffic flow and cannot fully utilize the highway capacity. Therefore, how to dynamically configure lanes and speed limits at highway toll plazas is still under study. Summary of the invention

[0005] In order to solve the technical problems existing in the above-mentioned background technology, the present invention provides a method and system for controlling the traffic status in the toll station area of ​​a highway. First, the short-term traffic demand of the toll station is predicted based on the LSTM neural network model. Then, the cellular transport model (CTM) is used to predict the traffic evolution in the toll station area. Furthermore, the MPC-based control framework is used to optimize the number of ETC and MTC lanes and the variable speed limit to minimize the total driving time of vehicles in the toll station area.

[0006] In order to achieve the above object, the present invention adopts the following technical solution:

[0007] A first aspect of the present invention provides a method for controlling traffic status in a highway toll station area.

[0008] A method for controlling traffic status in a highway toll station area, comprising:

[0009] Obtain traffic data from toll booths and upstream gantries;

[0010] Based on the traffic data of the toll station and the upstream gantries, the LSTM neural network model is used to predict the traffic demand of the toll station within a period of time;

[0011] Based on the traffic demand, a cellular transmission model is adopted to obtain the traffic flow operation state of the toll station within a period of time, and the traffic flow operation state includes traffic volume and traffic density;

[0012] Based on the traffic flow operation state and the number of lanes, an objective function for the total driving time of vehicles in the toll station area is constructed. Combining with the constraint conditions, model predictive control is used for iterative optimization calculation to obtain the speed change limit and lane configuration corresponding to the minimum total driving time.

[0013] Further, the cellular transmission model determines the traffic volume by identifying the change in cell density, where the change in cell density is:

[0014]

[0015] where ρ i (t) represents the single-lane traffic flow density of cell i at the initial moment of the t-th time interval; q i (t) represents the traffic volume entering cell i from cell i - 1 within the t-th time interval; q i+1 (t) represents the traffic volume leaving cell i and entering cell i + 1 within the t-th time interval; Δt represents the time interval of the unit time step; λ i (t) is the number of lanes of cell i within the t-th time interval.

[0016] Furthermore, the cellular transmission model determines the traffic volume of the main road cell by the minimum value of the number of vehicles sent out by cell i - 1 within the t-th time interval and the number of vehicles received by cell i within the t-th time interval;

[0017] The cellular transmission model determines the traffic volume of the merging cell according to the magnitude relationship between the traffic volume of cell i - 1 located on the main line and the traffic volume of cell i located on the ramp entering the downstream merging cell i;

[0018] The cellular transmission model determines the traffic volume of the diverging cell according to the following formula:

[0019]

[0020] where R or,i (t) represents the traffic volume that the exit ramp cell i can receive.

[0021] The second aspect of the present invention provides a traffic state control system for a highway toll station area.

[0022] A traffic state control system for a highway toll station area includes:

[0023] A data acquisition module, which is configured to: acquire traffic data of the toll station and the upstream gantry;

[0024] A traffic demand prediction module, which is configured to: based on the traffic data of toll stations and upstream gantries, adopt an LSTM neural network model to predict the traffic demand of a toll station for a period of time;

[0025] A traffic evolution prediction module, which is configured to: based on the traffic demand, adopt a cellular transmission model to obtain the traffic flow operation state of a toll station for a period of time, and the traffic flow operation state includes traffic volume and traffic density;

[0026] A control module, which is configured to: based on the traffic flow operation state and the number of lanes, construct an objective function for the total driving time of vehicles in the toll station area, combine constraint conditions, adopt model predictive control, and perform rolling and iterative optimization calculations to obtain the speed change limit and lane configuration corresponding to the minimum total driving time.

[0027] The third aspect of the present invention provides a computer-readable storage medium.

[0028] A computer-readable storage medium, on which a computer program is stored, and when the program is executed by a processor, it implements the steps in the traffic state control method for a highway toll station area as described in the first aspect above.

[0029] The fourth aspect of the present invention provides a computer device.

[0030] A computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the steps in the traffic state control method for a highway toll station area as described in the first aspect above.

[0031] Compared with the prior art, the beneficial effects of the present invention are:

[0032] First, the present invention predicts the short-term traffic demand of a toll station based on an LSTM neural network model. Then, the cellular transmission model (CTM) is applied to predict the traffic evolution in the toll station area. Further, the ETC and MTC lane numbers and variable speed limits are optimized using an MPC-based control framework to minimize the total driving time of vehicles in the toll station area. Finally, microscopic simulation experiments are carried out in VISSIM to verify the effectiveness of the proposed control method. The present invention integrates dynamic lane configuration and variable speed limit control (VSL) to alleviate congestion at toll stations. The simulation results show that the control method can significantly improve the traffic performance of toll stations and enhance the traffic efficiency of highways. The research results can guide the on-site implementation of active control at toll stations.

[0033] The control method established by the present invention can dynamically propose reasonable and effective lane configuration and variable speed limit adjustment schemes according to the actual traffic state, which can improve the overall traffic performance of toll stations and alleviate traffic congestion at toll stations.

[0034] The cellular transmission model adopted by the present invention has a high prediction accuracy when predicting the traffic evolution at toll stations. BRIEF DESCRIPTION OF THE DRAWINGS

[0035] The accompanying drawings forming a part of the present invention are used to provide a further understanding of the present invention. The schematic embodiments of the present invention and their descriptions are used to explain the present invention and do not constitute an improper limitation to the present invention.

[0036] Figure 1 is a framework diagram of a traffic state control method for highway toll station areas;

[0037] Figure 2 is a schematic diagram of the CTM model;

[0038] Figure 3 is a simulation modeling diagram;

[0039] Figure 4 is a simulation flow chart;

[0040] Figure 5 is a traffic demand diagram in microscopic simulation;

[0041] Figure 6(a1) is a flow distribution diagram between Cell 1 in CTM and microscopic simulation;

[0042] Figure 6(a2) is a density distribution diagram between Cell 1 in CTM and microscopic simulation;

[0043] Figure 6(b1) is a flow distribution diagram between Cell 2 in CTM and microscopic simulation;

[0044] Figure 6(b2) is a density distribution diagram between Cell 2 in CTM and microscopic simulation;

[0045] Figure 6(c1) is a flow distribution diagram between Cell 3 in CTM and microscopic simulation;

[0046] Figure 6(c2) is a density distribution diagram between Cell 3 in CTM and microscopic simulation;

[0047] Figure 6(d1) is a flow distribution diagram between Cell4 in CTM and microscopic simulation;

[0048] Figure 6(d2) is a density distribution diagram between Cell 4 in CTM and microscopic simulation;

[0049] Figure 6(e1) is a flow distribution diagram between Cell 5 in CTM and microscopic simulation;

[0050] Figure 6(e2) is a density distribution diagram between Cell 5 in CTM and microscopic simulation;

[0051] Figure 6 (f1) is the flow distribution diagram between Cell 6 in CTM and microscopic simulation;

[0052] Figure 6 (f2) is the density distribution diagram between Cell 6 in CTM and microscopic simulation;

[0053] Figure 6 (g1) is the flow distribution diagram between Cell 7 in CTM and microscopic simulation;

[0054] Figure 6 (g2) is the density distribution diagram between Cell 7 in CTM and microscopic simulation;

[0055] Figure 6 (h1) is the flow distribution diagram between Cell 8 in CTM and microscopic simulation;

[0056] Figure 6 (h2) is the density distribution diagram between Cell 8 in CTM and microscopic simulation;

[0057] Figure 7 (a1) is the trend diagram of flow rate change before and after the control of Cell 0;

[0058] Figure 7 (a2) is the trend diagram of speed change before and after the control of Cell 0;

[0059] Figure 7 (a3) is the trend diagram of density change before and after the control of Cell 0;

[0060] Figure 7 (b1) is the trend diagram of flow rate change before and after the control of Cell 1;

[0061] Figure 7 (b2) is the trend diagram of speed change before and after the control of Cell 1;

[0062] Figure 7 (b3) is the trend diagram of density change before and after the control of Cell 1;

[0063] Figure 7 (c1) is the trend diagram of flow rate change before and after the control of Cell 2;

[0064] Figure 7 (c2) is the trend diagram of speed change before and after the control of Cell 2;

[0065] Figure 7 (c3) is the trend diagram of density change before and after the control of Cell 2;

[0066] Figure 7 (d1) is the trend diagram of flow rate change before and after the control of Cell 3;

[0067] Figure 7 (d2) is the trend diagram of speed change before and after the control of Cell 3;

[0068] Figure 7 (d3) is the trend diagram of density change before and after the control of Cell 3;

[0069] Figure 7 (e1) is the trend chart of the traffic flow change before and after the control of Cell 4;

[0070] Figure 7 (e2) is the trend chart of the speed change before and after the control of Cell 4;

[0071] Figure 7 (e3) is the trend chart of the density change before and after the control of Cell 4;

[0072] Figure 7 (f1) is the trend chart of the traffic flow change before and after the control of Cell 5;

[0073] Figure 7 (f2) is the trend chart of the speed change before and after the control of Cell 5;

[0074] Figure 7 (f3) is the trend chart of the density change before and after the control of Cell 5;

[0075] Figure 7 (g1) is the trend chart of the traffic flow change before and after the control of Cell 6;

[0076] Figure 7 (g2) is the trend chart of the speed change before and after the control of Cell 6;

[0077] Figure 7 (g3) is the trend chart of the density change before and after the control of Cell 6;

[0078] Figure 7 (h1) is the trend chart of the traffic flow change before and after the control of Cell 7;

[0079] Figure 7 (h2) is the trend chart of the speed change before and after the control of Cell 7;

[0080] Figure 7 (h3) is the trend chart of the density change before and after the control of Cell 7;

[0081] Figure 7 (i1) is the trend chart of the traffic flow change before and after the control of Cell 8;

[0082] Figure 7 (i2) is the trend chart of the speed change before and after the control of Cell 8;

[0083] Figure 7 (i3) is the trend chart of the density change before and after the control of Cell 8;

[0084] Figure 8 (a) is the control effect diagram of the lane configuration control plan;

[0085] Figure 8 (b) is the control effect diagram of the variable speed limit control plan;

[0086] Figure 9 is the change chart of the average queue length of the MTC lane;

[0087] Figure 10 is the change chart of the service imbalance coefficient;

[0088] Among them,Figure 6(a1)-Figure 6(h2) The dashed line in the figure represents microscopic simulation, and the solid line represents CTM; Figure 7(a1)-Figure 7(i3) The dashed line in the figure represents the situation before control, and the solid line represents the situation after control. Specific implementation manners

[0089] The present invention will be further described below in conjunction with the accompanying drawings and embodiments.

[0090] It should be noted that the following detailed description is illustrative and is intended to provide further explanation of the present invention. Unless otherwise specified, all technical and scientific terms used in this embodiment have the same meaning as commonly understood by those of ordinary skill in the technical field to which the present invention belongs.

[0091] It should be noted that the terms used herein are only for describing specific implementation manners and are not intended to limit the exemplary embodiments according to the present invention. As used herein, unless the context clearly indicates otherwise, the singular form is also intended to include the plural form. In addition, it should be understood that when the terms "comprise" and / or "include" are used in this specification, they specify the presence of features, steps, operations, devices, components, and / or combinations thereof.

[0092] It should be noted that the flowcharts and block diagrams in the accompanying drawings illustrate the possible architectures, functions, and operations of methods and systems according to various embodiments of the present disclosure. It should be noted that each block in the flowchart or block diagram may represent a module, a program segment, or a part of code, and the module, program segment, or part of code may include one or more executable instructions for implementing the logical functions specified in each embodiment. It should also be noted that in some alternative implementations, the functions marked in the blocks may occur in a different order from that marked in the accompanying drawings. For example, two consecutive blocks shown may actually be executed substantially in parallel, or they may sometimes be executed in the reverse order, depending on the functions involved. Similarly, it should be noted that each block in the flowchart and / or block diagram, as well as the combinations of blocks in the flowchart and / or block diagram, may be implemented using a dedicated hardware-based system for performing the specified functions or operations, or may be implemented using a combination of dedicated hardware and computer instructions.

[0093] Embodiment 1

[0094] This embodiment provides a traffic status control method for highway toll station areas. This embodiment takes the application of this method to a server as an example. It can be understood that this method can also be applied to terminals, and can also be applied to a system including terminals and servers, and is realized through the interaction between terminals and servers. The server can be an independent physical server, or a server cluster or distributed system composed of multiple physical servers, or a cloud server that provides basic cloud computing services such as cloud services, cloud databases, cloud computing, cloud functions, cloud storage, web servers, cloud communications, middleware services, domain name services, security services CDN, and big data and artificial intelligence platforms. The terminal can be a smart phone, a tablet computer, a laptop computer, a desktop computer, a smart speaker, a smart watch, etc., but is not limited thereto. The terminal and the server can be directly or indirectly connected through wired or wireless communication methods, and this application does not make any restrictions here. In this embodiment, the method includes the following steps:

[0095] Obtain traffic data of the toll station and the upstream gantry;

[0096] Based on the traffic data of the toll station and the upstream gantry, use the LSTM neural network model to predict the traffic demand of the toll station for a period of time;

[0097] Based on the traffic demand, use the cell transmission model to obtain the traffic flow operation state of the toll station for a period of time, and the traffic flow operation state includes traffic volume and traffic density;

[0098] Based on the traffic flow operation state and the number of lanes, construct an objective function for the total driving time of vehicles in the toll station area, and combine the constraint conditions. Use model predictive control to perform rolling and iterative optimization calculations to obtain the speed limit and lane configuration corresponding to the minimum total driving time.

[0099] The specific solution of this embodiment can be implemented by the following content:

[0100] This embodiment designs a congestion control method for the toll station area based on model predictive control (MPC) to obtain the optimal control strategy. Model predictive control is a model-based closed-loop optimization control strategy. Its core idea is to use the prediction model to predict the dynamic characteristics of the system in the future for a period of time (i.e., the prediction period) in each control cycle, and use the rolling strategy to perform iterative optimization calculations, so that the uncertainties caused by model errors, etc., can be compensated, and then seek the optimal control strategy for the finite time domain in the current control cycle.

[0101] Specifically, this control method collects highway traffic flow data, predicts the traffic evolution at highway toll stations, and controls the traffic flow by using upstream variable speed limits and dynamic lane configurations at toll stations. First, install data collectors at the toll stations under study to record traffic data; second, based on the historical and real-time traffic data of the toll stations and upstream gantries, identify the traffic flow operation status, and use the LSTM neural network model to predict the short-term traffic demand at the toll stations; further, input the predicted traffic demand into the traffic state estimation model, i.e., the Cell Transmission Model (CTM), to obtain the future traffic evolution under control; finally, optimize the future traffic state by solving the objective function to achieve the optimal speed limit and lane configuration. Its control process is as Figure 1 shown.

[0102] This embodiment constructs a traffic demand prediction model based on the vehicle passing data of upstream toll stations, gantries, and the toll stations under study. First, screen the historical vehicle passing data of upstream toll stations, gantries, and the toll stations under study from the database, and reconstruct the data set by using the sliding window segmentation method; on this basis, use the processed historical data to train and test the LSTM model, and establish a traffic demand prediction model for the toll stations under study. The LSTM model is a neural network model with the ability to remember long-term and short-term information, and has good results in predicting time series data, so it can be applied to the prediction of traffic data. Finally, with the help of the traffic demand prediction model, use the real-time traffic data of the toll stations and upstream gantries to periodically predict (15 minutes in this embodiment) the traffic demand at the toll stations, providing a data basis for subsequent traffic state discrimination.

[0103] As Figure 2 shown, the Cell Transmission Model is an approximation of the first-order discrete Godunov model. In CTM, a road is discretized into section units of length L i (i = 1, 2,..., I), and each road unit contains at most one exit and one entrance. At the same time, time is divided into time intervals of interval Δt (t = 1, 2,..., T). In this embodiment, the last cell I represents the toll station.

[0104] Traffic flow density dynamics model

[0105] Traffic density is one of the representative parameters for identifying traffic states. In CTM, the change in cell density can be expressed as:

[0106]

[0107] In the formula, ρ i (t) represents the single-lane traffic flow density of cell i at the initial moment of the t-th time interval, veh / km / ln; q i$(t)$ represents the traffic flow from cell $i - 1$ into cell $i$ within the $t$-th time interval, in veh / h; $q$ i+1 $(t)$ represents the traffic flow leaving cell $i$ and entering cell $i + 1$ within the $t$-th time interval, in veh / h; $\Delta t$ represents the time interval of a unit time step, in h; $\lambda$ i $(t)$ is the number of lanes of cell $i$ within the $t$-th time interval, in ln.

[0108] Boundary flow dynamics model

[0109] The main road can be divided into three types of cells: main road cells, merging cells, and diverging cells. For main road cells, the traffic flow $q$ i $(t)$ can be represented by the minimum value of the number of vehicles emitted from cell $i - 1$ ($S$ i-1 $(t)$) and the number of vehicles received by cell $i$ ($R$ i $(t)$) within the $t$-th time interval, as shown in Equation (2). $S$ i-1 $(t)$ is the smaller value between the number of vehicles in cell $i - 1$ and the traffic capacity of cell $i - 1$ within the $t$-th time interval, as shown in Equation (3). Similarly, $R$ i $(t)$ is the smaller value between the remaining traffic capacity of cell $i$ and the traffic capacity of cell $i$ within the $t$-th time interval, as shown in Equation (4).

[0110] $q$ i $(t)=\min\{S$ i-1 $(t),R$ i $(t)\} (2)$

[0111] $S$ i-1 $(t)=\min\{v$ f $\cdot\rho$ i-1 $(t)\cdot\lambda$ i-1 $(t),Q$ i $\cdot\lambda$ i-1 $(t)\} (3)$

[0112] $R$ i $(t)=\min\{\omega\cdot(\rho$ jam $-\rho$ i $(t))\cdot\lambda$ i $(t),Q$ i $\cdot\lambda$ i $(t)\} (4)$

[0113] In the formula, $v$ f represents the free flow speed, in km / h; $Q$ represents the traffic capacity of cell $i$, in veh / h / ln; $\omega$ represents the congestion wave propagation speed, in km / h; $\rho$ jam represents the single - lane block density when the traffic flow speed is 0, in veh / km / ln.

[0114] For the merging cells, the traffic volume of cell i-1 on the main line and cell i on the ramp entering the downstream merging cell i can be calculated according to the traffic capacity of the downstream cell. If the acceptance capacity of the downstream cell is sufficient, the traffic flow can be calculated according to formula (5).

[0115]

[0116] In the formula, the mid{·} function represents taking the median value; θ i represents the merging coefficient, usually taken as the ratio of the traffic capacity of the on-ramp cell to the upstream main-line cell; S or，i (t) represents the traffic flow emitted by ramp cell i, in veh / h.

[0117] For the diverging cells, vehicles enter the off-ramp cell i from the main-line cell i-1 at a ratio of θ i . At this time, the vehicles entering the downstream main-line cell i can be regarded as: vehicles of different toll types enter the ETC or MTC toll stations respectively. Therefore, cell i can be regarded as a diverging cell. At the same time, the traffic density ρ I (t) of the ETC toll station and the traffic density ρ r，I (t) of the MTC toll station can be calculated according to formula (1).

[0118]

[0119] In the formula, R or，i (t) represents the traffic flow that off-ramp cell i can receive.

[0120] Model modification considering variable speed limits

[0121] When the vehicle is driving in the free-flow state, the desired speed of the driver is the free-flow speed of the road section. At this time, if a variable speed limit is set, the driver will reduce the vehicle speed to comply with the speed limit requirement, and at the same time, the maximum traffic capacity of the road section is restricted by the variable speed limit value and decreases; as the traffic density increases, the vehicle speed decreases, and the driving desired speed will be lower than the speed limit value, that is, when the variable speed limit is set on the road section, the vehicle driving speed is the smaller value of the desired speed and the variable speed limit. The sending capacity S i (t) and the receiving capacity R i (t) of the cell i controlled by the variable speed limit are as follows:

[0122] S i (t) = min{V i (t)·ρ i (t)·λ i (t), Q i，VSL (t)·λ i (t)} (7)

[0123] R i (t) = min{ω·(ρ jam-ρ i (t))·λ i (t), Q i，VSL (t)·λ i (t)} (8)

[0124] In the formula, V i (t) represents the driving speed of the vehicle in the t-th time interval under variable speed limit, km / h; Q i，VSL (t) represents the single-lane passing capacity of cell i in the t-th time interval under variable speed limit, veh / h / ln.

[0125] Based on the actual microscopic traffic conditions in the toll station and main line areas, this embodiment divides the main line section into different cells, and at the same time collects the flow, speed, and density data of each cell in the congestion period and off-peak period scenarios for parameter calibration and determination of the basic diagram of the three traffic flow parameters. Further, by comparing the calibrated CTM model with the corresponding key parameters in the actual traffic scenario, the accuracy of the parameters and the effectiveness of the model are verified.

[0126] The Mean Absolute Percentage Error (MAPE) is selected as the model effectiveness verification index. MAPE is the mean of the ratio of the difference between the model prediction and the actual result to the actual value. The closer the MAPE value is to 0%, the closer the model is to perfection. When the MAPE value is lower than 10%, it can be considered that the result predicted by the CTM model is the same as the actual traffic scenario. The MAPE calculation is as follows:

[0127]

[0128] In the formula, represents the predicted density value of cell i at the t-th time step, veh / km / ln; ρ i (t) represents the actual density value of cell i at the t-th time step, veh / km / ln; is the predicted flow value of cell i at the t-th time step, veh / h; q i (t) is the actual flow value of cell i at the t-th time step, veh / h.

[0129] Objective function

[0130] This embodiment takes the total travel time (TTT) as the optimization objective function TTT represents the total time used by the vehicle to pass through the toll station area, h.

[0131]

[0132] In the formula, u = (V, λ I , λr，I ); λ I represents the number of lanes for MTC, ln; λ r，I represents the number of lanes for ETC, ln; TTT is calculated as follows:

[0133]

[0134] Constraint conditions

[0135] (1) Service imbalance coefficient between ETC and MTC lanes

[0136] In the case where the arrival ratios of ETC and MTC vehicles differ significantly, to meet the control objective of minimizing the total travel time, dynamic lane configuration may increase the lanes corresponding to the vehicle type with a larger proportion without limit, resulting in a significant difference in the service levels of the two types of lanes. To reduce the operating cost and ensure the balance of the toll collection level of the toll station lanes, this embodiment defines the service imbalance coefficient ζ as a constraint condition for optimization control, and the calculation method is as follows:

[0137]

[0138] In the formula, ζ max represents the maximum value of the service imbalance coefficient (in this embodiment, ζ max = 10).

[0139] (2) Range of change in the number of lanes

[0140] In the actual operation of the highway, the real-time lane conversion between ETC and MTC often involves the configuration of highway toll facilities and personnel. If the change in dynamic lane configuration is large each time, it will lead to an increase in toll operation costs, and more importantly, it will increase the difficulty of operation management. To reduce the impact of lane dynamic control on toll costs and avoid unsafe driving behaviors of drivers due to frequent lane type changes in the toll plaza area, the number of lanes allowed to change in each control cycle should be constrained, and the number of lane changes should meet the following constraints:

[0141] |λ I (t) - λ I (t + 1)| ≤ Δλ max (13)

[0142] |λ r，I (t + 1) + λ I (t + 1)| = λ total (14)

[0143] In the formula, Δλ max represents the maximum value of the range of change in the number of lanes (in this embodiment, Δλ max = 2); λ total represents the total number of lanes at the toll station.

[0144] (3) Spatiotemporal variation of variable speed limits

[0145] To avoid drastic acceleration and deceleration behaviors of drivers within a short period, the variable speed limit values in adjacent cycles should satisfy certain restrictions to ensure the driving comfort and safety of drivers. Therefore, in this embodiment, the variable speed limits will be constrained from both the time and space aspects.

[0146] |V i (t) - V i (t + 1)| ≤ ΔV max (15)

[0147] |V i (t + 1) - V i-1 (t + 1)| ≤ ΔV′ max (16)

[0148] In the formula, V i (t) represents the traffic flow speed limit value of cell i at the t-th time interval; ΔV max and ΔV′ max represent the maximum values of the spatiotemporal variation of the variable speed limit (in this embodiment, (ΔV max = ΔV′ max = 10 km / h).

[0149] In this embodiment, to verify the effectiveness of the proposed control method, a simulation test was conducted. The toll station area is 400 m long and has a total of 15 lanes, with 6 lanes set at the entrance and 9 lanes set at the exit. The toll booth is 40 m long, and a variable message sign is set above it to dynamically display the lane types. In addition, microwave radars and cameras are also set in the toll station area to track vehicle trajectories, identify the road traffic conditions, and simultaneously predict the traffic flow and vehicle composition of the toll station. Data in August 2021 showed that the morning and evening traffic peaks on weekdays start at 8:00 and 17:00, and on weekends start at 10:00 and 17:00. 93.1% of the vehicles are cars, and 71.4% of the vehicles are ETC vehicles.

[0150] Using VISSIM software, this embodiment modeled a 2.7-kilometer-long highway. According to the on-site investigation, the static speed limit of the ETC lane was set at 20 km / h, and the static speed limit of the upstream main line was set at 60 km / h. The entire highway was divided into 11 Cells from Cell 0 to Cell 10 according to the road geometric characteristics, the cell length was set from 50 m to 500 m, and the time interval Δt was set at 6 s to ensure the continuous operation of vehicles in all cells.

[0151] To implement variable speed limit control, Cell1, 3, and 5 are selected as the control cells for variable speed limit, which are located 1 km, 1.8 km, and 2.5 km upstream of the toll station respectively. When implementing variable speed limit, the variable speed limit is optimized at Cell5, and the speed is gradually limited at Cell1 and Cell3. At the same time, the upper limit of the variable speed limit value is set to 60 km / h, the lower limit is set to 20 km / h, and the speed change between variable speed limit control cells does not exceed 10 km / h.

[0152] Before on-site application, a microscopic simulation test was conducted on the proposed control method in this embodiment. First, based on the road geometric characteristics and the collected traffic data, a microscopic simulation model was established using VISSIM software; in the microscopic simulation, the field measurement data of Cell0 was used as the actual traffic demand of the road network; the CTM model used the 20-minute traffic flow prediction based on historical data and the LSTM model as the traffic demand. Then, the simulated vehicles started from the upstream main line and the entrance ramp, entered the highway test section, paid the toll at the toll booth, and flowed into the downstream highway. To truly reflect the operation of the traffic flow, confluence and divergence behaviors were defined in the corresponding sections in this embodiment. Further, a hard separation between lanes was set at the toll booth, and vehicles equipped with ETC and those without ETC could only drive on their respective ETC or MTC lanes. According to the field investigation, the speed limit of the ETC lane was set to 20 km / h, and it passed through the toll booth without stopping; while the MTC lane needed to stop for toll payment, and the service time was set to 23 seconds. Finally, the microscopic simulation model was calibrated using the field traffic data to improve the accuracy of the model.

[0153] During the simulation process, to change the speed limit value and the toll lane type, it is necessary to call the VISSIM Component Object Model (COM) Application Programming Interface (API). The overall process is as Figure 4 shown. First, use a Python program to call the VISSIM API to load the traffic network, start the simulation, and optimize the control variable u. Then, use the API to collect traffic data, transfer the data to the Python optimization program, and at the same time input the returned optimal control into VISSIM for iterative loop. When the lane configuration changes, the Python program deletes the original toll type through the VISSIM API and modifies the toll type on the corresponding lane. At the same time, the drivable vehicle type of the corresponding lane is updated. When the speed limit changes, the Python program adds the required speed decision points in specific sections and updates the speed limit through the VISSIM API.

[0154] In terms of traffic flow input, according to the historical data during the evening peak period, the traffic demand for each period in the simulated road network was determined (see Figure 5 ), and the total simulation period was 135 minutes. The first 15 minutes of the simulation period ( Figure 5(-15 minutes to 0 minutes), the traffic flow gradually enters and fills the road network. In the subsequent 120-minute simulation, the scenarios before and after control are compared to evaluate the control performance of the integrated control method.

[0155] Finally, within the prediction time horizon (N p = 15 min), the lane configuration and speed limit sequence are solved by minimizing the objective function. The genetic algorithm is used to solve the variable speed limit value and the optimal lane configuration, and the variable speed limit value is updated every 5 minutes and the lane configuration is updated every 10 minutes.

[0156] Before using CTM as the prediction model of the MPC framework, this embodiment first calibrates the parameters in CTM. Traffic data collectors are placed at the beginning of each cell of the microscopic simulation. The model parameters of free flow speed v_f, capacity Q_i, shock wave speed ω, and jam density ρ_jam are corrected to minimize the difference between the CTM results and the simulation data. Then, this study verifies the corrected CTM and obtains a MAPE of 9.2%. The value of MAPE and Figure 6(a1)-Figure 6(h2) the traffic flow and density profiles in it show that the corrected CTM can accurately predict traffic evaluations in free flow and congestion situations and can perform active control.

[0157] Before using CTM as the prediction model of the MPC framework, this embodiment first calibrates the parameters in CTM. By setting data collection points at the starting position of each cell, traffic data such as traffic volume, speed, and density of each cell are obtained. At the same time, for the free flow speed v f and traffic capacity Q i and congestion wave speed ω and jam density ρ jam and other model parameters are corrected to minimize the difference between the prediction results of CTM and the simulation data. This embodiment verifies the corrected CTM model and obtains an average absolute percentage error (MAPE) of 9.8%. The results show that the corrected CTM model can accurately predict traffic states in congestion and free flow situations and can be applied to active traffic control.

[0158] The control method is implemented through microscopic simulation, and the changes in traffic flow characteristics under the conditions of the uncontrolled scheme and the MPC-based congestion control scheme are collected to evaluate the actual effect of the control method. Figure 7(a1)-Figure 7(i3) The changes in traffic flow characteristics in terms of flow, speed, and density before and after control are shown. Figure 8(a) is the control effect diagram of the lane configuration, and Figure 8(b) is the control effect diagram of the variable speed limit. From Figure 7(a1)-Figure 7(i3)It can be seen that Cell0 - Cell4, located upstream of the main line, are less affected by bottleneck congestion before and after control, and the change trends of traffic flow before and after control are less different; for Cell 5 - Cell 8, the traffic flow differences before and after control are not significant during the period of gentle traffic flow changes. During the period of obvious traffic flow changes, the control measures can significantly increase the traffic flow of the bottleneck section (Cell 7, Cell8) and reduce the impact on the vehicles on the main line (Cell 5, Cell 6).

[0159] After implementing the control scheme, since vehicles must stop or decelerate on Cell7 and Cell8, the average speeds of the two cells are basically the same as before control, and the impact of implementing the control scheme on these two cells is limited. The traffic speeds of Cell5 and Cell6 after control are significantly higher than those before control, indicating that the control method can effectively alleviate the phenomenon of congestion propagation upstream in the bottleneck area. In addition, due to the impact of variable speed limits, the speeds of the upstream cells (Cell3 and Cell4) are lower within 30 minutes after implementing the control. However, due to avoiding the occurrence of congestion, the overall average speed is higher than before control. Similarly, the traffic speeds of Cell1 and Cell2 after control are slightly lower than before control. During the simulation, the average density of the overall road section in the research area decreased from 79.30 veh / km / ln in the uncontrolled scenario to 38.50 veh / km / ln, and the maximum cell density decreased from 192.40 veh / km / ln to 187.37 veh / km / ln.

[0160] In addition, the queue lengths before and after control are measured, and the change in the average queue length of the MTC lane is obtained as Figure 9 shown. It can be seen from the figure that the proposed control method shortens the average queue length during the simulation process by 40.1% (from 69.68 m to 41.72 m).

[0161] The imbalance coefficients before and after control are calculated according to formula (12), as Figure 10 shown. During the simulation process, the imbalance coefficient after control is always less than that before control, and even during a short - term congestion, the imbalance coefficient remains stable without significant increase. Further, to quantify the fluctuation range of the imbalance coefficient, the coefficient of variation (the ratio of the standard deviation to the mean) is introduced. The results show that the standard deviations of the imbalance coefficients before and after control are 33.17 and 8.67 respectively, the means are 19.41 and 7.16 respectively, and the coefficients of variation are 1.71 and 1.21 respectively. In summary, after implementing the control scheme, the service levels between the ETC and MTC lanes are more balanced

[0162] According to Equation (12), Figure 10This is the change in the imbalance coefficient before and after the experiment. The imbalance coefficient after control is always less than that before control. Even when congestion is triggered, the imbalance coefficient remains stable without significant growth. To measure the fluctuation of the imbalance coefficient, the coefficient of variation, which is the ratio of its standard deviation to the mean, is used. The simulation results show that the standard deviations of the imbalance coefficients before and after control are 33.17 and 8.67 respectively, and the means are 19.41 and 7.16 respectively. Therefore, the coefficient of variation before control is 1.71, and the coefficient of variation after control is 1.21. In summary, after the implementation of control, the imbalance coefficient between lanes is significantly reduced and the fluctuation is more stable, and the service levels between ETC and MTC lanes are more balanced.

[0163] 4.5 Statistical Test

[0164] To further prove the effectiveness of the proposed control method, this study ran 10 simulations by changing the random seed number. Table 1 shows the paired t-test results of all 10 simulation results. Assuming that the simulation results follow a normal distribution and the significance level is 0.05. The results in Table 1 prove that the control method proposed in this embodiment significantly reduces the TTT and the average speed, thus reducing the congestion at the toll plaza.

[0165] Table 1 t-Test Results

[0166]

[0167]

[0168] Embodiment 2

[0169] This embodiment provides a traffic state control system for highway toll stations.

[0170] A traffic state control system for highway toll stations includes:

[0171] A data acquisition module configured to: acquire traffic data of toll stations and upstream gantries;

[0172] A traffic demand prediction module configured to: based on the traffic data of toll stations and upstream gantries, use an LSTM neural network model to predict the traffic demand of toll stations for a period of time;

[0173] Based on the traffic demand, use a cellular transmission model to obtain the traffic flow operation state of toll stations for a period of time, where the traffic flow operation state includes traffic volume and traffic density;

[0174] A control module, which is configured to: construct an objective function for the total driving time of vehicles in the toll station area based on the traffic flow operation state and the number of lanes, and in combination with constraint conditions, adopt model predictive control, perform rolling and iterative optimization calculations, and obtain the speed change limit and lane configuration corresponding to the minimum total driving time.

[0175] It should be noted here that the above data acquisition module, traffic demand prediction module, traffic evolution prediction module, and control module are the same as the examples and application scenarios implemented in the steps corresponding to those in Embodiment 1, but are not limited to the content disclosed in Embodiment 1 above. It should be noted that the above modules, as part of the system, can be executed in a computer system such as a set of computer-executable instructions.

[0176] Embodiment 3

[0177] This embodiment provides a computer-readable storage medium, on which a computer program is stored. When the program is executed by a processor, it implements the steps in the traffic state control method for the highway toll station area as described in Embodiment 1 above.

[0178] Embodiment 4

[0179] This embodiment provides a computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor. When the processor executes the program, it implements the steps in the traffic state control method for the highway toll station area as described in Embodiment 1 above.

[0180] Those skilled in the art should understand that the embodiments of the present invention can be provided as a method, a system, or a computer program product. Therefore, the present invention can be implemented in the form of a hardware embodiment, a software embodiment, or an embodiment combining software and hardware aspects. Moreover, the present invention can be implemented in the form of a computer program product implemented on one or more computer-usable storage media (including but not limited to disk storage and optical storage, etc.) containing computer-usable program code.

[0181] The present invention is described with reference to the flowcharts and / or block diagrams of methods, apparatuses (systems), and computer program products according to embodiments of the present invention. It should be understood that each flow and / or block in the flowchart and / or block diagram, and the combination of flows and / or blocks in the flowchart and / or block diagram, can be implemented by computer program instructions. These computer program instructions can be provided to the processor of a general-purpose computer, a special-purpose computer, an embedded processor, or other programmable data processing devices to generate a machine, such that the instructions executed by the processor of the computer or other programmable data processing devices generate means for implementing the functions specified in Figure 1 one process or multiple processes and / or blocks Figure 1 one block or multiple blocks.

[0182] These computer program instructions can also be stored in a computer-readable memory that can direct a computer or other programmable data processing device to work in a specific manner, such that the instructions stored in the computer-readable memory produce a manufacture including an instruction device that implements the functions specified in one process or a plurality of processes and / or one block or a plurality of blocks in the flow Figure 1 in one process or a plurality of processes and / or Figure 1 one block or a plurality of blocks.

[0183] These computer program instructions can also be loaded onto a computer or other programmable data processing device, such that a series of operation steps are performed on the computer or other programmable device to produce a computer-implemented process, and thus the instructions executed on the computer or other programmable device provide steps for implementing the functions specified in one process or a plurality of processes and / or Figure 1 one block or a plurality of blocks. Figure 1 in one block or a plurality of blocks.

[0184] Those of ordinary skill in the art can understand that all or part of the processes in the methods of the above embodiments can be completed by instructing relevant hardware through a computer program. The program can be stored in a computer-readable storage medium. When the program is executed, it can include the processes of the embodiments of the above methods. Among them, the storage medium can be a magnetic disk, an optical disk, a read-only memory (ROM), or a random access memory (RAM), etc.

[0185] The above are only the preferred embodiments of the present invention and are not intended to limit the present invention. For those skilled in the art, the present invention can have various changes and modifications. Any modification, equivalent replacement, improvement, etc. made within the spirit and principle of the present invention shall be included in the protection scope of the present invention.

Claims

1. A traffic state control method for highway toll station areas, characterized in that, it includes: Obtain traffic data of the toll station and the upstream gantry; Based on the traffic data of the toll station and the upstream gantry, use the LSTM neural network model to predict the traffic demand of the toll station for a period of time; Based on the traffic demand, use the cell transmission model to obtain the traffic flow operation state of the toll station for a period of time, and the traffic flow operation state includes traffic volume and traffic density; Based on the traffic flow operation state and the number of lanes, construct an objective function for the total driving time of vehicles in the toll station area, combine the constraint conditions, and use model predictive control to perform rolling and repeated optimization calculations to obtain the speed change limit and lane configuration corresponding to the minimum total driving time; The cell transmission model determines the traffic volume by identifying the change in cell density, where the change in cell density is: Among them, represents the single-lane traffic flow density of the cell at the initial moment of the th time interval; The traffic flow entering cell from cell within the th time interval is represented by ; The traffic flow leaving cell and entering cell within the th time interval is represented by ; represents the time interval of a unit time step; is the number of lanes of cell within the The objective function is: Among them, represents the objective function; TTT represents the total driving time of the vehicle in the toll station area; , 、 both represent the predicted density value of the cell; ; represents the number of lanes of the MTC, represents the number of lanes of the ETC, and V represents the variable speed limit value; represents the th time step cell 's predicted density value.

2. The traffic state control method for highway toll station areas according to claim 1, characterized in that, The cellular transmission model determines the traffic flow of the main road cells through the minimum value of the number of vehicles sent by the cells within the th time interval and the number of vehicles received by the cells within the th time interval; th time interval and the number of vehicles received by the cells within the th time interval. The cellular transmission model determines the traffic flow of the merging cell according to the magnitude relationship of the traffic flows of the cells located on the main line and the cells located on the ramp entering the downstream merging cell ; The cell transmission model determines the traffic volume of the diversion cell according to the following formula: Among them, represents the exit ramp cell the acceptable traffic flow; represents the number of vehicles received by the cell within the th time interval; is the number of vehicles sent by the cell within the th time interval.

3. The traffic state control method for highway toll station areas according to claim 1, characterized in that, The constraint conditions include the service imbalance coefficient : Among them, represents the maximum value of the service imbalance coefficient.

4. The traffic state control method for highway toll station areas according to claim 1, characterized in that, The constraint conditions include the number of lane changes: Among them, represents the maximum change amplitude value of the number of lanes; represents the total number of lanes of the toll station.

5. The traffic state control method for highway toll station areas according to claim 1, characterized in that, The constraint conditions include the spatio-temporal change of variable speed limits: Among them, represents the traffic flow speed limit value of the th time interval cell; represents the maximum value of the variable speed limit time change; represents the maximum value of the variable speed limit space change.

6. A traffic state control system for highway toll station areas, characterized in that, it includes: A data acquisition module configured to: obtain traffic data of the toll station and the upstream gantry; A traffic demand prediction module configured to: based on the traffic data of the toll station and the upstream gantry, use the LSTM neural network model to predict the traffic demand of the toll station for a period of time; A traffic evolution prediction module configured to: based on the traffic demand, use the cell transmission model to obtain the traffic flow operation state of the toll station for a period of time, and the traffic flow operation state includes traffic volume and traffic density; A control module configured to: based on the traffic flow operation state and the number of lanes, construct an objective function for the total driving time of vehicles in the toll station area, combine the constraint conditions, and use model predictive control to perform rolling and repeated optimization calculations to obtain the speed change limit and lane configuration corresponding to the minimum total driving time; The cell transmission model determines the traffic volume by identifying the change in cell density, where the change in cell density is: Among them, represents the single-lane traffic density of the cell at the initial moment of the th time interval; The traffic flow density of the single lane of the cell at the initial moment of the represents the traffic volume entering cell from cell during the th time interval; represents the traffic volume leaving cell and entering cell during the th time interval; represents the time interval of a unit time step; is the number of lanes of cell during the th time interval; The objective function is: Among them, represents the objective function; TTT represents the total driving time of the vehicle in the toll station area; , 、 both represent the predicted density value of the cell; ; represents the number of lanes of the MTC, represents the number of lanes of the ETC, and V represents the variable speed limit value; represents the th time step cell 's predicted density value.

7. A computer-readable storage medium, on which a computer program is stored, characterized in that, When the program is executed by a processor, it implements the steps in the traffic state control method for highway toll station areas according to any one of claims 1-5.

8. A computer device, including a memory, a processor, and a computer program stored on the memory and executable on the processor, characterized in that, When the processor executes the program, it implements the steps in the traffic state control method for highway toll station areas according to any one of claims 1-5.

Images