A coordinated control method for expressway ramp control and variable speed limit considering accident risk

Abstract

The present invention relates to a method for collaborative control of expressway ramps and variable speed limits taking into account accident risks. The purpose of the present invention is to solve the limitation problem of existing collaborative control methods that unilaterally pursue the improvement of traffic efficiency while ignoring the risk and impact of accidents. The process is: 1. Obtain the road conditions of the target expressway section; 2. Deploy detectors on each section and each ramp; 3. Determine the area of action of collaborative control; 4. Establish a traffic flow model; 5. Calibrate the parameters of the traffic flow model; 6. Obtain real-time traffic data in each control cycle; 7. Determine whether collaborative control is turned on; 8. Calculate the average collision probability in the area of action of collaborative control and determine the risk status; 9. Obtain the optimal ramp control adjustment rate and speed limit value; 10. Enter the next control cycle and execute a new round of optimization control again from step 6. The present invention belongs to the field of expressway safety control.

Description

Technical Field

[0001] The present invention belongs to the field of expressway safety control, and in particular relates to an expressway ramp control and a variable speed limit coordinated control method taking accident risks into consideration. Background Art

[0002] Expressways form the backbone of urban road networks, and their operational efficiency is closely linked to the overall efficiency of urban transportation. However, with the surge in urban transportation demand, expressways are facing increasingly severe traffic congestion, and their advantages of speed, efficiency, and safety are gradually being eroded. Therefore, there is an urgent need to improve traffic management to better utilize the existing expressway network and fully tap the potential of existing transportation facilities. Ramp control and variable speed limit control are the main control methods for expressways, but the effectiveness of each control method is limited and it is difficult to adapt to the increasing traffic pressure. The coordinated control of ramp control and variable speed limit has the potential to combine the control effects of both and further improve the current situation of expressway traffic congestion.

[0003] Existing collaborative control methods are largely based on international experience with expressway control, primarily designed to alleviate bottleneck congestion on expressways. They unilaterally pursue optimized traffic efficiency, with insufficient attention paid to road safety. Expressways have numerous entrance and exit ramps, often creating recurring congestion bottlenecks. These bottlenecks create a complex and volatile traffic environment, leading to significant safety issues. Furthermore, due to the dense distribution of ramps, high vehicle entry and exit frequency, and high traffic volumes on expressways, a traffic accident can easily involve multiple adjacent ramps. Ramp overflows can further impact connected urban roads, potentially paralyzing the entire road network. Therefore, preventing and controlling expressway accident risks is crucial for expressway management and control.

[0004] In summary, in view of the limitations of existing collaborative control methods that unilaterally pursue the improvement of traffic efficiency while ignoring the risk and impact of accidents, it is necessary to establish a collaborative control method for expressway ramp control and variable speed limit that takes accident risks into consideration. Summary of the Invention

[0005] The purpose of this invention is to solve the limitation problem of existing collaborative control methods that unilaterally pursue the improvement of traffic efficiency while ignoring the risk and impact of accidents, and to propose an expressway ramp control and variable speed limit collaborative control method that takes accident risks into consideration.

[0006] A method for coordinated control of expressway ramps and variable speed limits considering accident risks is characterized in that the method comprises the following steps:

[0007] Step 1: Obtain the road conditions of the target expressway section, including the total length L of the section, the number of lanes m, and the location, number, and spacing of entrance and exit ramps along the section;

[0008] Step 2: Based on the road conditions of the target expressway section obtained in Step 1, the target section is divided into N segments, numbered 1, 2, ..., N, and detectors are deployed in each segment of the main line and on each ramp;

[0009] Step 3: Based on Step 1 and Step 2, determine the collaborative control action area;

[0010] Step 4: Based on Step 1, Step 2, and Step 3, a traffic flow model adapted to cooperative control is established;

[0011] Step 5: Use detectors deployed in each section to obtain expressway flow, density, and speed data to calibrate traffic flow model parameters;

[0012] Step 6: In each control cycle m, obtain the real-time traffic data of each detector on the expressway, including the flow rate q of each section i (k), speed v i (k), density ρ i (k) and the ramp queue length w i (k);

[0013] Among them, q i (k), v i (k),ρ i (k) represents the flow, velocity and density of segment i at time k, w i (k) represents the queue length of entrance ramp i at the current time k;

[0014] Step 7: Calculate the average traffic flow speed V in the coordinated control area determined in step 3. avg The traffic congestion status is determined and based on this, whether the coordinated control is enabled is determined. The specific process is as follows:

[0015] Calculate the average traffic flow speed V in the cooperative control area using detector data avg , judge V avg Whether the speed threshold V is exceeded threshold If yes, it is determined to be a free flow state, the variable speed limit control is not performed in the current cycle, and the ramp control resumes the maximum adjustment rate; otherwise, it is determined to be a congested state and proceed to step 8;

[0016] Step 8: Calculate the average collision probability CP in the cooperative control action area avg And determine the risk status and decide whether to conduct routine prevention and control of accident risks or focus on prevention and control;

[0017] The specific process is:

[0018] The collision probability model is used to predict the real-time collision probability of each segment and calculate the average collision probability CP of the cooperative control area. avg, as the predicted value of accident risk;

[0019] Determine the average collision probability CP avg Has the preset collision probability threshold CP been reached? threshold If it is determined to be high risk, a collaborative control optimization model will be constructed with safety as the sole control objective to focus on accident risk prevention and control; otherwise, if it is determined to be low risk, a collaborative control optimization model will be constructed with efficiency and safety as the control objectives to conduct routine accident risk prevention and control;

[0020] Step 9: Solve the constructed collaborative control optimization model to obtain the optimal ramp control adjustment rate r i (m) and speed limit VSL i (m), sent to the corresponding control device for execution;

[0021] Step 10: After the control cycle m ends, the next control cycle m+1 is entered, and a new round of optimization control is performed again from step 6.

[0022] The beneficial effects of the present invention are:

[0023] The purpose of the present invention is to address the limitation of existing collaborative control technology that unilaterally pursues traffic efficiency without fully considering the characteristics of frequent accidents and large traffic involved on expressways, and to propose a collaborative control method for expressway ramp control and variable speed limit that takes accident risks into consideration.

[0024] 1. The present invention predicts the real-time accident risk of expressways by introducing a collision probability model, flexibly determines the control target of collaborative control according to the size of the accident risk in each control cycle, and can achieve real-time monitoring and flexible prevention and control of accident risk status.

[0025] 2. The present invention divides traffic conditions according to average speed and flexibly determines the opening and closing of cooperative control, which conforms to the complex and changeable traffic characteristics of urban expressways and can better avoid the negative impact of meaningless cooperative control when traffic conditions are good.

[0026] 3. This invention fully considers the complex and volatile nature of expressway traffic environments and the dynamic changes in accident risk. It utilizes common loop detectors to proactively monitor traffic conditions and predict risk profiles, thereby achieving a reasonable balance between safety and efficiency in its control objectives. By fully integrating the advantages of ramp control and variable speed limits, it effectively achieves comprehensive improvements in expressway traffic conditions. BRIEF DESCRIPTION OF THE DRAWINGS

[0027] Figure 1 This is an example of segment division and detector arrangement according to the present invention;

[0028] Figure 2 This is the basic framework of expressway collaborative control described in the present invention;

[0029] Figure 3 It is a control flow chart of the expressway ramp control and variable speed limit coordinated control method considering accident risks described in the present invention. DETAILED DESCRIPTION

[0030] Specific implementation method 1: The specific process of the expressway ramp control and variable speed limit coordinated control method considering accident risks in this implementation method is as follows:

[0031] Step 1: Obtain the road conditions of the target expressway section, including the total length L of the section, the number of lanes m, and the location, number, and spacing of entrance and exit ramps along the section;

[0032] Based on the uncontrolled traffic congestion situation, the target expressway section that needs to be controlled is determined. The total length L of the target section, the number of lanes m, and the location, number, and spacing of entrance and exit ramps along the line are calculated to prepare for the layout of detectors, the installation of control facilities, and the establishment of traffic flow models.

[0033] Step 2: Based on the road conditions of the target expressway section obtained in Step 1, the target section is discretized (discrete the entire road length into small sections) into N segments, numbered 1, 2, ..., N, based on the principle of continuity and uniformity of road attributes (dividing a whole road into small sections, while maintaining the same attributes such as the number of lanes in each section). Detectors are deployed on each segment and each ramp of the main line (expressways and highways are generally composed of ramps and main roads, with the main road being the main line) to obtain real-time traffic flow data such as flow rate q, density ρ, and speed v for each segment and each ramp.

[0034] Examples of road segment division and detector placement are as follows: Figure 1 As shown;

[0035] Step 3: Based on Step 1 and Step 2, determine the collaborative control action area;

[0036] Step 4: Based on steps 1, 2, and 3, considering the respective mechanisms of ramp control and variable speed limit as well as the multiple congestion bottlenecks of the expressway, a macroscopic traffic flow model for the expressway that adapts to coordinated control is established.

[0037] Based on the classic METANET model, adaptive improvements are made to the model considering the driver's speed limit compliance level, reaction time characteristics, the structural characteristics of multiple congestion bottlenecks on expressways, and the mechanism of ramp control, to establish a freeway macro traffic flow model that is suitable for freeway collaborative control scenarios.

[0038] Step 5: Use detectors deployed in each section to obtain expressway flow, density, and speed data to calibrate the parameters of the macro traffic flow model;

[0039] Step 6: In each control cycle m, obtain the real-time traffic data of each detector on the expressway, including the flow rate q of each section i (k), speed v i (k), density ρ i (k) and the ramp queue length w i (k) etc.

[0040] Among them, q i (k), v i (k),ρ i (k) represents the flow, velocity and density of segment i at time k, w i (k) represents the queue length of the entrance ramp i at the current time k. The above real-time traffic data will be used to determine the traffic state and be used as the initial state of the traffic flow model.

[0041] Step 7: Calculate the average traffic flow speed V in the coordinated control area determined in step 3. avg And determine the traffic congestion status, and based on this, determine whether to start the coordinated control; the specific process is as follows:

[0042] Calculate the average traffic flow speed V in the cooperative control area using detector data avg , judge V avg Whether the speed threshold V is exceeded threshold If yes, it is determined to be a free flow state, the variable speed limit control is not performed in the current cycle, and the ramp control resumes the maximum adjustment rate; otherwise, it is determined to be a congested state and proceed to step 8;

[0043] Step 8: Calculate the average collision probability CP in the cooperative control action area avg And determine the risk status and decide whether to conduct routine prevention and control of accident risks or focus on prevention and control;

[0044] The specific process is:

[0045] Based on the real-time detected traffic flow parameters, the collision probability model is used to predict the real-time collision probability of each segment, and the average collision probability CP of the cooperative control area is calculated. avg , as the predicted value of accident risk;

[0046] Afterwards, determine the average collision probability CP avg Has the preset collision probability threshold CP been reached? thresholdIf it is determined to be high risk, a collaborative control optimization model will be constructed with safety as the sole control objective to focus on accident risk prevention and control; otherwise, if it is determined to be low risk, a collaborative control optimization model will be constructed with efficiency and safety as the control objectives to conduct routine accident risk prevention and control;

[0047] Note that both focused and routine accident risk prevention and control are based on the model predictive control (MPC) framework, which uses traffic flow models to predict future traffic flow states under various control schemes, and selects the control scheme that achieves the optimal control target as the optimal solution.

[0048] Step 9: Solve the constructed collaborative control optimization model to obtain the optimal ramp control adjustment rate r i (m) and speed limit VSL i (m), sent to the corresponding control equipment (ramp signal light, variable information board) for execution;

[0049] Due to the specific nature of the model, solving the optimization model is relatively simple. For a given speed limit scenario, the optimal ramp adjustment rate is automatically determined during the iterative prediction process of the traffic flow model. Therefore, the decision variables of the optimization model are essentially the speed limits for each segment and control cycle within the prediction time domain. Speed limits are a series of discrete values and are subject to strict constraints. Therefore, the number of feasible solutions to the optimization model is limited, and can be solved through traversal enumeration.

[0050] Step 10: After the control cycle m ends, the next control cycle m+1 is entered, and a new round of optimization control is performed again starting from step 6.

[0051] Specific embodiment 2: This embodiment differs from specific embodiment 1 in that each segment in step 2 includes at most one entrance ramp and one exit ramp;

[0052] The segment division should not be too long or too short, usually between 300m and 1000m;

[0053] The main line needs to have a set of detectors in each section and each lane at the beginning of the section, such as Figure 1 As shown in LD0~LD8;

[0054] The on-ramp needs to be equipped with two sets of detectors, namely entry detectors and queue detectors, to obtain the actual influx of the ramp and the queue length respectively;

[0055] The exit ramp only needs to be equipped with one set of detectors to obtain the ramp exit volume.

[0056] Other steps and parameters are the same as those in the first embodiment.

[0057] Specific embodiment three: This embodiment differs from specific embodiment one or two in that, in step 3, the coordinated control action area is determined based on step 1 and step 2; the specific process is:

[0058] In light of the actual traffic flow, variable speed limit information boards (display boards that can display speed limit information) are laid out on the main line to form several variable speed limit sections;

[0059] Ramp signals are installed at entrance ramps that are prone to traffic bottlenecks (according to the experience of traffic managers, entrance ramps that are prone to mainline congestion are selected) to form several ramp control nodes;

[0060] The road area where the speed limit section and ramp control node operate is the collaborative control area.

[0061] Other steps and parameters are the same as those in the first or second embodiment.

[0062] Specific embodiment 4: This embodiment differs from any one of specific embodiments 1 to 3 in that, in step 4, based on steps 1, 2, and 3, the respective mechanisms of ramp control and variable speed limit, as well as the multiple congestion bottlenecks of the expressway, are considered to establish a macroscopic traffic flow model for the expressway that adapts to the coordinated control.

[0063] Based on the classic METANET model, adaptive improvements are made to the model considering the driver's speed limit compliance level, reaction time characteristics, the structural characteristics of multiple congestion bottlenecks on expressways, and the mechanism of ramp control, to establish a freeway macro traffic flow model that is suitable for freeway collaborative control scenarios.

[0064] The specific process is:

[0065] 1. The classic METANET model is a recursive relationship between the flow, speed, and density of each road segment over time, as follows:

[0066]

[0067] Among them, q i (k), v i (k),ρ i (k) represents the flow, velocity and density of segment i at time k; w i (k) represents the queue length of entrance ramp i at time k, h i (k) represents the on-ramp flow rate of segment i at time k, in veh / h; s i (k) represents the outgoing flow of the exit ramp at time segment i in veh / h; λ i is the number of lanes in segment i; v f is the free flow speed in km / h; ρcrit is the critical density, unit is veh / (km·ln); μ, η, κ, τ are all model parameters; T is the time step; L i is the length of segment i; V[ρ i (k)] is the density ρ i (k) The corresponding expected speed.

[0068] 2. Based on the METANET model in 1, improvements were made to the model in consideration of driver speed limit compliance, reaction time characteristics, the structural characteristics of multiple congestion bottlenecks on expressways, and the mechanism of ramp control. The resulting macroscopic traffic flow model for expressways (expressways are urban roads built within cities with central dividers, full access control, controlled entrance and exit spacing and type, one-way dual lanes or more, and equipped with supporting traffic safety and management facilities) is as follows (Formulas 3-11 together form the model):

[0069] (1) Dynamic density equation:

[0070]

[0071]

[0072] w i (k) = w i (k-1)+T·[d i (k-1)-h i (k-1)] (5)

[0073] r i (k) = r i (k-1)+K R [O crit,i -O i (k-1)] (6)

[0074] Where, d i (k) and w i (k) represents the ramp arrival flow and the number of vehicles queuing on the ramp at time k for the entrance ramp corresponding to segment i;

[0075] Q i is the maximum capacity of the entrance ramp contained in segment i;

[0076] ρ jam,i and ρ crit,i are the blocking density and critical density of the ramp segment i respectively;

[0077] Formula (6) is the ALINEA ramp control algorithm. i (k) is the ramp regulation rate of segment i at time k, in veh / h;

[0078] O crit,i is the critical occupancy of segment i;

[0079] O i (k-1) is the actual occupancy of segment i at time k-1;

[0080] K R To adjust the parameters;

[0081] (2) Dynamic velocity equation:

[0082]

[0083]

[0084] Where, τ i (k) represents the reaction time parameter of segment i at time k;

[0085] τ dec Indicates the reaction time under deceleration state;

[0086] τ uni Indicates the reaction time under constant speed;

[0087] τ acc Indicates the reaction time under accelerated state;

[0088] (3) Expected speed equation:

[0089] V[ρ i (k)]=min{v f ·exp{-(1 / μ)[ρ i (k) / ρ crit ] μ},(1+α i (k))·u i vsl (k)} (9)

[0090]

[0091] Where, α i (k) represents the real-time speeding amplitude of segment i at time k, which is used to reflect the driver's violation of the given speed limit and is calculated using the simple moving average method;

[0092] represents the displayed speed limit value of segment i at time kt;

[0093] T αThe number of moving average time periods (the simple moving average method uses the average of the values of the past few sampling periods as an estimate of the future value. For example, if I use the average of the speed of the past 5 sampling periods to calculate the future speed, then the "number of moving average time periods" is 5);

[0094] (4) Section flow equation:

[0095] q i (k) = λ i ρ i (k)v i (k) (11).

[0096] The other steps and parameters are the same as those in the first to third embodiments.

[0097] Specific embodiment 5: This embodiment differs from specific embodiments 1 to 4 in that in step 5, detectors arranged in each section are used to obtain expressway flow, density, and speed data to calibrate the parameters of the macro traffic flow model; the specific process is as follows:

[0098] The macro traffic flow model parameters include traffic flow basic parameters and macro traffic flow model global parameters;

[0099] (1) Obtain a flow-density scatter plot using a detector;

[0100] Based on the flow-density scatter plot, the basic traffic flow parameters are obtained. The basic traffic flow parameters include the free flow speed v f , critical density ρ crit , blocking density ρ jam wait;

[0101] (2) Using the optimization algorithm, the global parameter calibration model with the goal of minimizing the prediction error of flow and speed is solved to obtain the global parameters of the macroscopic traffic flow model;

[0102] The specific process is:

[0103] The objective function of the global parameter calibration model with the goal of minimizing the prediction error of flow and speed is:

[0104]

[0105] The constraints of the global parameter calibration model with the goal of minimizing the prediction error of flow and speed are:

[0106] σ min ≤σ≤σ max

[0107] σ min =[0.001,0.001,0.001,10,10]

[0108] σ max =[0.05,0.05,0.05,60,60]

[0109] τ acc -τ dec >0

[0110] Among them, σ is the global parameter vector of the macroscopic traffic flow model, that is, [τ dec ,τ uni ,τ acc ,η,κ],τ dec Indicates the reaction time in deceleration state, τ uni represents the reaction time under uniform speed, τ acc represents the reaction time under acceleration, η is the model parameter, κ is the model parameter, σ min is the minimum value of the global parameter vector of the traffic flow model, σ max is the maximum value of the global parameter vector of the traffic flow model;

[0111] N is the total number of road discrete segments; K is the maximum time step of time discretization; v i,real (k) is the actual average velocity of segment i at time k; v i,predict (k|σ) is the average speed predicted by the macroscopic traffic flow model of segment i under the global parameter vector σ at time k; q i,real (k) is the actual flow rate of segment i at time k; q i,predict (k|σ) is the predicted flow of the macro traffic flow model of segment i under the global parameter vector σ at time k;

[0112] The objective function is solved using intelligent optimization algorithms such as genetic algorithms to obtain the optimal global parameters of the macro traffic flow model.

[0113] Establish the basic framework of expressway coordinated control based on model predictive control (MPC) and design basic control parameters, including sampling period T, prediction time domain N p , control time domain N c wait;

[0114] The specific steps are:

[0115] The basic framework of expressway coordinated control based on model predictive control (MPC) is established; the specific process is as follows:

[0116] The control object of the present invention is the expressway system (the entire expressway can be regarded as a traffic system). Based on the basic principles and ideas of MPC, the following is established: Figure 2 The basic framework of collaborative control is shown in Figure 1. The control principle is:

[0117] (1) At the beginning of each control cycle, the expressway system obtains the current traffic status, including flow, speed, density, etc.

[0118] (2) Subsequently, the collected traffic flow status information is input into the traffic flow model as the initial condition for predicting the future traffic flow status of the expressway system. This process is performed in each rolling optimization, thereby achieving feedback correction of the model;

[0119] (3) Based on the traffic flow model and the customized optimization objective, multi-step predictions are performed and the objective function is calculated under different control schemes (the combination of ramp control signals and variable speed limits is a control scheme). The optimal control scheme (ramp adjustment rate and speed limit) is obtained by solving the optimization problem in real time.

[0120] (4) Finally, the optimal control solution is fed back to the METANET model and used together with the traffic flow status as the initial condition for the next control cycle; at the same time, it is also sent to the expressway system for actual execution.

[0121] Design basic control parameters; the specific process is:

[0122] (1) Sampling period T: It represents the time interval for the control system to obtain the state of the controlled object. To ensure the predictive significance of the prediction model, the sampling period should ensure that there is enough time for sampling when the vehicle is traveling at the fastest speed, that is, it satisfies:

[0123] T≤min(L i / v f ) (1)

[0124] (2) Prediction time domain N p : Refers to the length of time that the prediction model predicts the controlled object. It should not be too short or too long, and is generally an integer multiple of the sampling period;

[0125] (3) Control time domain N c : Refers to the time length of the control sequence to predict future traffic conditions, solve the optimal control solution and return it to the system for execution. It should also be an integer multiple of the sampling period. N c It should be determined in combination with the cycles of each control signal to ensure that at least one ramp control cycle and variable speed limit cycle are executed;

[0126] The collaborative control method proposed in this invention is based on the MPC control framework. Under this framework, the important control parameters of the collaborative control are designed. After that, the actual control stage is entered. The collaborative control of the expressway is carried out according to steps 6 to 10. The complete control process is as follows: Figure 3 shown.

[0127] The other steps and parameters are the same as those in the first to fourth embodiments.

[0128] Specific embodiment 6: This embodiment differs from any one of the specific embodiments 1 to 5 in that the average collision probability CP of the cooperative control action area is calculated in step 8. avg And determine the risk status, and decide whether to conduct routine accident risk prevention and control or key prevention and control; the specific process is:

[0129] Step 8.1: Calculate the real-time collision probability of each segment and the average collision probability CP of the cooperative control action area avg ;

[0130] Step 8.2: Determine the current average collision probability CP avg Whether the collision probability threshold CP is exceeded threshold If yes, it is determined to be high risk and go to step 8.3; if no, it is determined to be low risk and go to step 8.4;

[0131] Step 8.3: Build a collaborative control optimization model (Formula 16) with safety as the sole control objective to focus on accident risk prevention and control;

[0132] Step 8.4: Take efficiency and safety as control objectives to construct a collaborative control optimization model (Formula 27) to conduct routine prevention and control of accident risks.

[0133] The other steps and parameters are the same as those in the first to fifth embodiments.

[0134] Specific embodiment 7: This embodiment differs from any one of specific embodiments 1 to 6 in that the real-time collision probability of each segment and the average collision probability CP of the cooperative control action area are calculated in step 8.1. avg ; The specific process is:

[0135] The real-time collision probability of each segment can be calculated based on the flow, speed, and occupancy data collected by the detector. The calculation formula is:

[0136]

[0137] Where, P collision (SV, AO, SS) represents the real-time collision probability given the traffic characteristics SV, AO, and SS. SV is the standard deviation of traffic flow in the period 5-10 minutes before the target time, AO is the average occupancy rate in the period 0-5 minutes before the target time, and SS is the standard deviation of speed in the period 5-10 minutes before the target time. b0, b1, b2, and b3 are model parameters with values of -3.694, -1.207, 3.149, and 4.028, respectively.

[0138] The above formula is used to obtain the real-time collision probability CP of each segment i in the cooperative control action area: i After that, the average collision probability CPavg It can be calculated as:

[0139]

[0140] Where M is the total number of segments included in the collaborative control action area, and C is the numbered set of segments included in the collaborative control action area.

[0141] The other steps and parameters are the same as those in the first to sixth embodiments.

[0142] Specific embodiment eight: This embodiment differs from any one of specific embodiments one to seven in that in step 8.3, a collaborative control optimization model is constructed with safety as the only control objective to focus on accident risk prevention and control; the specific process is as follows:

[0143] The sum of the real-time collision probabilities of each segment at each time step in the prediction time domain is calculated as the quantitative value of the overall accident risk level of the expressway in the future time step; specifically,

[0144] For any segment i, its real-time collision probability CP at time k is i (k) can be calculated as:

[0145]

[0146] Among them, SV i (k), AO i (k), SS i (k) represents the standard deviation of the flow rate of segment i in the 5-10 minutes before time k, the average occupancy in the first 0-5 minutes, and the standard deviation of the speed in the first 5-10 minutes.

[0147] The flow rate, occupancy rate and speed required for calculation are predicted by the METANET model. If there is no predicted result for the time range involved, the corresponding detector measured value is used to supplement it.

[0148] Therefore, the sum of the collision probabilities TCP of each time segment in the prediction time domain can be calculated as:

[0149]

[0150] where N p is the prediction time domain, N is the total number of discrete road segments;

[0151] At this time, the expressway is in a high-risk state, so the control goal is to minimize the total collision probability. The objective function of the constructed collaborative control optimization model is:

[0152] J = minTCP (16)

[0153] Where J is the objective function;

[0154] The constraints are:

[0155] (1) Within the same control cycle, the speed limit value of any segment at each time step remains unchanged;

[0156]

[0157] Among them, VSL i (m) represents the speed limit value of segment i within the control period m, T c Indicates the length of the control cycle, represents the speed limit value of segment i at time k;

[0158] (2) The variable speed limit value of any control cycle shall not exceed the maximum speed limit standard, nor be lower than the minimum speed limit standard;

[0159] V min ≤VSL i (m)≤V max (18)

[0160] Among them, V min is the minimum speed limit standard, V max The maximum speed limit standard;

[0161] (3) To ensure the stable operation of the main line traffic flow, the change of speed limit values between adjacent control cycles should not be too large, and the difference of variable speed limit values between adjacent sections in the same control cycle should not be too large;

[0162] |VSL i (m+1)-VSL i (m)|≤10km / h (19)

[0163] |VSL i+1 (m)-VSL i (m)|≤20km / h (20)

[0164] Among them, VSL i (m+1) represents the speed limit value of segment i in control cycle m+1, VSL i (m) represents the speed limit value of segment i within control cycle m, VSL i+1 (m) represents the speed limit value of segment i+1 within control cycle m;

[0165] (4) Considering the driver's controllability of driving speed, the published variable speed limit value is set to a multiple of 10 km / h, that is, the variable speed limit value is selected from discrete values;

[0166] VSL i (m)∈{10km / h,20km / h,…,70km / h,80km / h} (21)

[0167] (5) According to basic traffic flow theory, the traffic flow of each section does not exceed the main line capacity;

[0168] λ i ρ i (k)v i (k)≤λ i Q max (twenty two)

[0169] Among them, Q max Indicates the main line capacity;

[0170] (6) To ensure the stability of ramp traffic flow, the regulation rate of each ramp should not fluctuate too much in time;

[0171]

[0172] The other steps and parameters are the same as those in the first to seventh embodiments.

[0173] Specific embodiment 9: This embodiment differs from specific embodiments 1 to 8 in that in step 8.4, efficiency and safety are both used as control objectives to construct a collaborative control optimization model to conduct routine prevention and control of accident risks; the specific process is as follows:

[0174] If the current average collision probability CP avg Does not exceed the collision probability threshold CP threshold , efficiency and safety are taken as control objectives together to build a collaborative control optimization model and implement routine prevention and control of accident risks.

[0175] At this time, the expressway is in a low-risk state, and collaborative control should take into account both safety and efficiency, with the control goals of improving expressway traffic efficiency and reducing collision risks.

[0176] In terms of traffic efficiency, the total travel time TTT and total mileage TTD within the predicted time domain are used as benefit indicators;

[0177] The calculations are as follows:

[0178]

[0179]

[0180] Among them, ρ i (k+j) represents the density of segment i at time (k+j), w i (k+j) represents the queue length of entrance ramp i at time (k+j), v i (k+j) represents the velocity of segment i at time (k+j);

[0181] In terms of safety, the total collision probability TCP is used as the benefit indicator, as shown in step 9.3, and is calculated as:

[0182]

[0183] The objectives of cooperative control are to minimize the total travel time TTT, maximize the total mileage TTD, and minimize the total collision probability TCP. Therefore, the linear weighted sum of TTT, TTD, and TCP is used as the objective function, that is:

[0184] J=min(α TTT TTT-α TTD TTD+α TCP TCP) (27)

[0185] Among them, α TTT , α TTD and α TCP are the weight coefficients of TTT, TTD and TCP respectively;

[0186] The constraints of the model are the same as those in step 8.3 and will not be repeated here.

[0187] The other steps and parameters are the same as those in Specific Embodiments 1 to 8.

[0188] The present invention may have many other embodiments. Without departing from the spirit and essence of the present invention, those skilled in the art may make various corresponding changes and modifications based on the present invention, but these corresponding changes and modifications should all fall within the scope of protection of the claims attached to the present invention.

Claims

1. A coordinated control method for expressway ramp control and variable speed limits that considers accident risks, characterized by: The specific process of the method is: Step 1: Obtain the road conditions of the target expressway section, including the total length L of the section, the number of lanes m, and the location, number, and spacing of entrance and exit ramps along the section; Step 2: Based on the road conditions of the target expressway section obtained in Step 1, the target section is divided into N segments, numbered 1, 2, ..., N, and detectors are deployed in each segment of the main line and on each ramp; Step 3: Based on Step 1 and Step 2, determine the collaborative control action area; Step 4: Based on Step 1, Step 2, and Step 3, a traffic flow model adapted to cooperative control is established; Step 5: Use detectors deployed in each section to obtain expressway flow, density, and speed data to calibrate traffic flow model parameters; Step 6: In each control cycle m, obtain the real-time traffic data of each detector on the expressway, including the flow rate q of each section i (k), speed v i (k), density ρ i (k) and the ramp queue length w i (k); Among them, q i (k), v i (k),ρ i (k) represents the flow, velocity and density of segment i at time k, w i (k) represents the queue length of entrance ramp i at the current time k; Step 7: Calculate the average traffic flow speed V in the coordinated control area determined in step 3. avg The traffic congestion status is determined and based on this, whether the coordinated control is enabled is determined. The specific process is as follows: Calculate the average traffic flow speed V in the cooperative control area using detector data avg , judge V avg Whether the speed threshold V is exceeded threshold If yes, it is determined to be a free flow state, the variable speed limit control is not performed in the current cycle, and the ramp control resumes the maximum adjustment rate; otherwise, it is determined to be a congested state and proceed to step 8; Step 8: Calculate the average collision probability CP in the cooperative control action area avg And determine the risk status and decide whether to conduct routine prevention and control of accident risks or focus on prevention and control; The specific process is: The collision probability model is used to predict the real-time collision probability of each segment and calculate the average collision probability CP of the cooperative control area. avg , as the predicted value of accident risk; Determine the average collision probability CP avg Has the preset collision probability threshold CP been reached? threshold If it is determined to be high risk, a collaborative control optimization model will be constructed with safety as the sole control objective to focus on accident risk prevention and control; otherwise, if it is determined to be low risk, a collaborative control optimization model will be constructed with efficiency and safety as the control objectives to conduct routine accident risk prevention and control; Step 9: Solve the constructed collaborative control optimization model to obtain the optimal ramp control adjustment rate r i (m) and speed limit VSL i (m), sent to the corresponding control device for execution; Step 10: After the control cycle m ends, the next control cycle m+1 is entered, and a new round of optimization control is performed again starting from step 6.

2. The expressway ramp control and variable speed limit coordinated control method considering accident risks according to claim 1 is characterized by: In step 2, each segment includes at most one entrance ramp and one exit ramp; A set of detectors is deployed in each lane of each section and the beginning of the section; The entrance ramp is equipped with two sets of detectors, namely an entry detector and a queue detector, to respectively obtain the actual influx of the ramp and the queue length; The exit ramp is provided with a group of detectors for obtaining the amount of people leaving the ramp.

3. The expressway ramp control and variable speed limit coordinated control method considering accident risks according to claim 2 is characterized by: In step 3, based on step 1 and step 2, the coordinated control action area is determined; the specific process is: Variable speed limit information boards are laid out on the main line to form several variable speed limit sections; Ramp signals are set up at the entrance ramps to form several ramp control nodes; The road area where the speed limit section and ramp control node operate is the collaborative control area.

4. The expressway ramp control and variable speed limit coordinated control method considering accident risks according to claim 3 is characterized by: In step 4, a traffic flow model adapted to cooperative control is established based on steps 1, 2, and 3. The specific process is as follows: The traffic flow model adapted to cooperative control is as follows: (1) Dynamic density equation: w i (k)=w i (k-1)+T·[d i (k-1)-h i (k-1)] (5) r i (k)=r i (k-1)+K R [O crit,i -O i (k-1)] (6) Where q i (k),ρ i (k) represents the flow and density of segment i at time k; h i (k) represents the on-ramp flow rate of segment i at time k, in veh / h; s i (k) represents the outgoing flow of the exit ramp at time segment i in veh / h; λ i is the number of lanes in segment i; L i is the length of segment i; T is the time step; ρ crit is the critical density, unit is veh / (km·ln); d i (k) and w i (k) represents the ramp arrival flow and the number of vehicles queuing on the ramp at time k for the entrance ramp corresponding to segment i; Q i is the maximum capacity of the entrance ramp contained in segment i; ρ jam,i and ρ crit,i are the blocking density and critical density of the ramp segment i respectively; r i (k) is the ramp regulation rate of segment i at time k, in veh / h; O crit,i is the critical occupancy of segment i; O i (k-1) is the actual occupancy of segment i at time k-1; K R To adjust the parameters; (2) Dynamic velocity equation: Where, τ i (k) represents the reaction time parameter of segment i at time k; τ dec Indicates the reaction time under deceleration state; τ uni Indicates the reaction time under constant speed; τ acc Indicates the reaction time under accelerated state; v i (k) represents the velocity of segment i at time k; V[ρ i (k)] is the density ρ i (k) the corresponding expected speed; η, κ are model parameters; (3) Expected speed equation: Where, α i (k) represents the real-time overspeed amplitude of segment i at time k; represents the displayed speed limit value of segment i at time kt; T α is the number of time periods for the moving average; v f is the free flow speed in km / h; μ is the model parameter; (4) Section flow equation: q i (k)=λ i r i (k)v i (k) (11).

5. The expressway ramp control and variable speed limit coordinated control method considering accident risks according to claim 4 is characterized by: In step 5, detectors deployed in each section are used to obtain data on expressway flow, density, and speed, and to calibrate traffic flow model parameters. The specific process is as follows: The traffic flow model parameters include traffic flow basic parameters and traffic flow model global parameters; (1) Obtain a flow-density scatter plot using a detector; Based on the flow-density scatter plot, the basic traffic flow parameters are obtained. The basic traffic flow parameters include the free flow speed v f , critical density ρ crit , blocking density ρ jam ; (2) Using the optimization algorithm, the global parameter calibration model with the goal of minimizing the prediction error of flow and speed is solved to obtain the global parameters of the traffic flow model; The specific process is: The objective function of the global parameter calibration model with the goal of minimizing the prediction error of flow and speed is: The constraints of the global parameter calibration model with the goal of minimizing the prediction error of flow and speed are: s min ≤σ≤σ max s min =[0.001,0.001,0.001,10,10] s max =[0.05,0.05,0.05,60,60] t acc -t dec >0 Among them, σ is the global parameter vector of the traffic flow model, that is, [τ dec ,τ uni ,τ acc ,η,κ]; τ dec Indicates the reaction time in deceleration state, τ uni represents the reaction time under uniform speed, τ acc represents the reaction time under acceleration, η is the model parameter, κ is the model parameter, σ min is the minimum value of the global parameter vector of the traffic flow model, σ max is the maximum value of the global parameter vector of the traffic flow model; N is the total number of road discrete segments; K is the maximum time step of time discretization; v i,real (k) is the actual average velocity of segment i at time k; v i,predict (k|σ) is the average speed predicted by the traffic flow model of segment i under the global parameter vector σ at time k; q i,real (k) is the actual flow rate of segment i at time k; q i,predict (k|σ) is the predicted flow of the traffic flow model of segment i under the global parameter vector σ at time k; The objective function is solved using the optimization algorithm to obtain the optimal global parameters of the traffic flow model.

6. The expressway ramp control and variable speed limit coordinated control method considering accident risks according to claim 5 is characterized by: In step 8, the average collision probability CP of the cooperative control action area is calculated. avg And determine the risk status and decide whether to conduct routine prevention and control of accident risks or focus on prevention and control; The specific process is: Step 8.1: Calculate the real-time collision probability of each segment and the average collision probability CP of the cooperative control action area avg ; Step 8.2: Determine the current average collision probability CP avg Whether the collision probability threshold CP is exceeded threshold If yes, it is determined to be high risk and go to step 8.3; if no, it is determined to be low risk and go to step 8.4; Step 8.3: Build a collaborative control optimization model with safety as the sole control objective to focus on accident risk prevention and control; Step 8.4: Take efficiency and safety as control objectives to build a collaborative control optimization model and conduct routine prevention and control of accident risks.

7. The expressway ramp control and variable speed limit coordinated control method considering accident risks according to claim 6 is characterized by: In step 8.1, the real-time collision probability of each segment and the average collision probability CP of the cooperative control action area are calculated. avg ; The specific process is: The real-time collision probability of each segment can be calculated based on the flow, speed, and occupancy data collected by the detector. The calculation formula is: Where, P collision (SV, AO, SS) represents the real-time collision probability under given traffic characteristics SV, AO, and SS, where SV is the standard deviation of the flow rate within 5-10 minutes before the target time, AO is the average occupancy rate within 0-5 minutes before the target time, and SS is the standard deviation of the speed within 5-10 minutes before the target time; b0, b1, b2, and b3 are model parameters; The real-time collision probability CP of each segment i in the cooperative control action area is obtained using formula (12): i After that, the average collision probability CP avg It can be calculated as: Where M is the total number of segments included in the collaborative control action area, and C is the numbered set of segments included in the collaborative control action area.

8. The expressway ramp control and variable speed limit coordinated control method considering accident risks according to claim 7 is characterized by: In step 8.3, a collaborative control optimization model is constructed with safety as the only control objective to focus on accident risk prevention and control; The specific process is: For any segment i, the real-time collision probability CP at time k i (k) is calculated as: Among them, SV i (k), AO i (k), SS i (k) represents the standard deviation of the flow rate in segment i in the 5-10 minutes before time k, the average occupancy in the first 0-5 minutes, and the standard deviation of the velocity in the first 5-10 minutes; Therefore, the sum of the collision probabilities TCP of each time segment in the prediction time domain can be calculated as: where N p is the prediction time domain, N is the total number of discrete road segments; At this time, the expressway is in a high-risk state, so the control goal is to minimize the total collision probability. The objective function of the constructed collaborative control optimization model is: J=min TCP (16) Where J is the objective function; The constraints are: (1) Within the same control cycle, the speed limit value of any segment at each time step remains unchanged; Among them, VSL i (m) represents the speed limit value of segment i within the control period m, T c Indicates the length of the control cycle, represents the speed limit value of segment i at time k; (2) The variable speed limit value of any control cycle shall not exceed the maximum speed limit standard, nor be lower than the minimum speed limit standard; In min ≤VSL i (m)=V max (18) Among them, V min is the minimum speed limit standard, V max The maximum speed limit standard; (3) To ensure the stable operation of main line traffic flow; |VSL i (m+1)-VSL i (m)|≤10km / h (19) |VSL i+1 (m)-VSL i (m)|≤20km / h (20) Among them, VSL i (m+1) represents the speed limit value of segment i in control cycle m+1, VSL i (m) represents the speed limit value of segment i within control cycle m, VSL i+1 (m) represents the speed limit value of segment i+1 within control cycle m; (4) The variable speed limit value is selected from discrete values; VSL i (m)∈{10km / h,20km / h,……,70km / h,80km / h} (21) (5) The flow rate of each section does not exceed the main line capacity; l i r i (k)v i (k)≤λ i Q max (22) Among them, Q max Indicates the main line capacity; (6) To ensure the stability of ramp traffic flow; 9. The expressway ramp control and variable speed limit coordinated control method considering accident risks according to claim 8 is characterized by: In step 8.4, efficiency and safety are taken as control objectives to construct a collaborative control optimization model to conduct routine prevention and control of accident risks. The specific process is as follows: In terms of traffic efficiency, the total travel time TTT and total mileage TTD within the predicted time domain are used as benefit indicators; The calculations are as follows: Among them, ρ i (k+j) represents the density of segment i at time (k+j); w i (k+j) represents the queue length of entrance ramp i at time (k+j); v i (k+j) represents the velocity of segment i at time (k+j); In terms of safety, the total collision probability TCP is used as the benefit indicator and is calculated as: The objectives of cooperative control are to minimize the total travel time TTT, maximize the total mileage TTD, and minimize the total collision probability TCP. Therefore, the linear weighted sum of TTT, TTD, and TCP is used as the objective function, that is: J=min(α TTT TTT-α TTD TTD+α TCP TCP) (27) Among them, α TTT , α TTD and α TCP are the weight coefficients of TTT, TTD and TCP respectively; The constraints are the same as in step 8.3.

Images