Unidirectional OD green wave path signal-oriented autonomous coordination and optimization method and system

Abstract

The invention discloses an autonomous coordination and optimization method for unidirectional OD green wave path signals. The method comprises the following steps: step 1, calculating an initial period of a single-point signal of an OD green wave path; step 2, carrying out self-adaptive timing of a single-point signal of an OD green wave path; and step 3, coordinating and optimizing OD green wave path signals. Compared with a traditional method needing manual intervention, the method achieves the full-autonomous process of signal coordination and optimization, remarkably improves the efficiency and accuracy of green wave band optimization, and reduces errors and delay caused by human factors.

Description

Technical Field

[0001] The present invention relates to the technical field of intelligent traffic management, and in particular to an autonomous coordination and optimization method and system for one-way OD green wave path signals. Background Art

[0002] Green wave signal coordination aims to design the phase difference between signals so that vehicles can pass through adjacent signal intersections in sequence without stopping, thereby improving the traffic efficiency of the road. The existing green wave coordination optimization models can be divided into one-way and two-way according to different application scenarios.

[0003] In the early signal coordination methods, the one-way coordination model is the most basic form of green wave control. Its core is to use the shortest travel time between adjacent signals as the phase difference to achieve the continuity of vehicle flow. However, with the complexity of traffic demand, the two-way coordination model has gradually become a research hotspot, such as MAXBAND, MULTI-BAND and AM-BAND models. These models introduce periodic constraints, speed constraints and other conditions, and combine mixed integer linear programming methods to seek the global optimal solution of bandwidth maximization to better meet the green wave control needs on two-way roads.

[0004] However, in order to more effectively respond to changes in travel demand from O (Origin) area to D (Destination) area, a new green wave path design method has emerged in recent years, namely the OD green wave path planning method. This method can dynamically adjust the one-way green wave path to adapt to changes in OD demand based on road network zoning, and realize the one-way green wave path planning from O area to D area. However, the existing green wave coordination method is difficult to directly apply to the signal coordination of the OD green wave path, and there are several major problems:

[0005] First, the traditional one-way coordination model cannot dynamically respond to changes in actual traffic flow due to the fixed green wave bandwidth, and it is difficult to meet the needs of future autonomous traffic signal control systems. Secondly, although some two-way coordination models can adapt to fluctuations in dynamic traffic flow, they are originally designed to serve two-way roads and cannot be directly applied to one-way OD green wave path scenarios. In addition, all existing green wave models assume that all traffic lights on the same path have the same cycle, and the calculation of the cycle requires an additional process, which introduces manual intervention in the coordination, making it difficult to achieve autonomous coordination of the green wave band.

[0006] In summary, the traditional green wave signal coordination method has obvious deficiencies in adaptability and dynamic adjustment capabilities, and it is difficult to meet the actual needs of OD green wave paths. These limitations significantly limit the application effect of OD green wave belts in future autonomous transportation systems, and they urgently need to be improved and perfected through technological innovation. Summary of the invention

[0007] In view of the above technical problems, the present invention provides an autonomous coordination and optimization method and system for unidirectional OD green wave path signals.

[0008] The present invention provides an autonomous coordination and optimization method for a unidirectional OD green wave path signal, comprising the following steps:

[0009] Step 1: Calculate the initial period of the single-point signal of the OD green wave path;

[0010] Step 2: When performing adaptive pre-configuration of a single-point signal on an OD green wave path;

[0011] Step 3: Coordinate and optimize OD green wave path signals.

[0012] Furthermore, step 2 specifically includes the following steps:

[0013] Step 2-1: Calculate the key quantity of phase alliance;

[0014] Step 2-2: Calculate the phase alliance characteristic function value;

[0015] Step 2-3: Calculate the Shapley value of the phase;

[0016] Step 2-4: Calculate the initial phase green light duration.

[0017] Furthermore, in step 2-1, the signal phase in the OD green wave path is regarded as a game participant, forming a cooperative game model G = [P, v]; where P is a phase alliance, the alliance combination mode is taken from the phase set N = {1, 2, 3, ..., n}, and v is the alliance characteristic function; the key quantity R i It represents the capacity ratio of phase i, and the calculation formula is:

[0018]

[0019] Among them, C cap,i represents the total traffic capacity (pcu / h) of all lanes allowed to be released by the single-point signal m in the i-th phase, and the starting loss time l of the phase is divided into the total green light duration T g The ratio of is introduced into the constraint condition of the key quantity; the constraint condition is as follows:

[0020]

[0021] Among them, |S| represents the number of participants in the phase alliance, and |S|≤n, l represents the phase starting loss time s, which is usually set to 2s; T g Indicates the initial common period C of the single-point signal m k,0The total green light duration (s) is calculated as follows: T g =C k,0 -A m -(z / u a -t s ).

[0022] Furthermore, in step 2-2, the constraint conditions of the key quantity are used to design the alliance characteristic function v(S) of the single-point signal pre-allocation model, and the calculation formula is as follows:

[0023]

[0024] This characteristic function follows a voting mechanism: if the sum of the capacity ratios of the alliance S satisfies the key quantity R i If the constraints are met, the alliance is effective; otherwise, its utility is zero.

[0025] Furthermore, in step 2-3, all phase alliances of each signal are listed Using the formula in step 2-2, calculate S for each alliance j The characteristic function value v(S j );Calculate the relationship between each phase i∈{1,2,3,…,n} in each phase alliance S j The average marginal contribution value in The calculation formula is:

[0026]

[0027] Where k is the number of phase alliances, n is the total number of signal phases, |S j |Indicates phase alliance S j The number of phases in the j ∪{i}) is the characteristic function value of the alliance containing phase i, v(S j ) is the characteristic function value of the alliance that does not contain phase i;

[0028] The average marginal contribution values ​​of all phases are combined into an n-dimensional Shapley vector It is used to guide the fair allocation of phase green light duration based on phase capacity.

[0029] Furthermore, in step 2-4, the average marginal contribution value of phase i obtained in step 2-3 is Multiply by the total green light duration T of the initial cycle g , get the initial green light duration of this phase The calculation formula is as follows:

[0030]

[0031] Among them, round means rounding operation to ensure that the green light duration is an integer second; after the rounding operation, the sum of the initial green light durations of all phases may be different from the original green light duration T g If they are not equal, the difference between the two is added to the phase with the longest green light duration, so as to finally determine the final initial green light duration of the single point signal m phase i. The calculation formula is as follows:

[0032]

[0033] If the sum of the initial green light durations of all phases equals the original green light duration T g If the difference is positive, the maximum green light phase duration increases; if the difference is negative, the maximum green light phase duration decreases; if the difference is zero, the maximum green light phase duration remains unchanged.

[0034] Furthermore, step three specifically includes the following steps:

[0035] Step 3-1: Prepare data;

[0036] Step 3-2: Set constraints;

[0037] Step 3-3: Design objective function;

[0038] Step 3-4: Process coordination optimization and results.

[0039] Further, step 3-2 includes the following steps:

[0040] Step 3-2-1: Setting of directional interference constraints

[0041] For adjacent signals m and m+1 in OD green wave path k, their direction constraints are:

[0042]

[0043] Among them, b m is the ratio of the green wave bandwidth between signal m and signal m+1 to the common period, w m is the interference variable between signal m and signal m+1, r m and r m+1 are the red light duration ratios of signal m and signal m+1, w m+1 is the interference variable between signal m+1 and signal m+2, τ m+1 Indicates the queue clearing duration ratio of signal m+1;

[0044] Step 3-2-2: Setting integer loop constraints

[0045] The one-way integer cycle constraints are:

[0046] w m -w m+1 +t m +δ m L m -δ m+1 L m+1 =(r m+1 -r m )+τ m+1 +q m ,

[0047]

[0048] Among them, t m represents the ratio of the travel time from signal m to signal m+1 to the period, δ m and δ m+1 0 / 1 variables representing the left turn phase pattern of signal m and signal m+1 respectively. If the left turn precedes the straight-ahead movement, the value is 1; otherwise, it is 0. m and L m+1 They represent the ratio of the left turn green light duration to the cycle of signal m and signal m+1, q m Represents a loop integer variable;

[0049] Step 3-2-3: Setting green wave speed constraints

[0050] In order to ensure that vehicles can reduce the number of stops or even not stop when traveling on the OD green wave path, it is necessary to set a one-way green wave speed constraint as follows:

[0051]

[0052] Among them, d m is the distance m between signal m and signal m+1, e m (f m ) are the upper and lower limits of the speed m, z is the frequency of the cycle (Hz);

[0053] Step 3-2-4: Setting green wave speed change constraints

[0054] Taking into account the positive and negative nature of speed change, the speed change constraint of the one-way green wave is:

[0055]

[0056] Among them, λ m The intermediate variable representing the linearization of the absolute value of the speed change of signal m, t m+1 represents the ratio of the travel time from signal m+1 to signal m+2 to the period, d m+1 is the distance m between signal m+1 and signal m+2, g m (h m) are the upper and lower limits of the speed change from signal m to signal m+1;

[0057] Step 3-2-5: Setting basic constraints on parameters and decision variables

[0058] The parameters and decision variables in the one-way direction from zone O to zone D in the present invention have the following constraints:

[0059]

[0060] Among them, b m is the ratio of the green wave bandwidth between signal m and signal m+1 to the common period, z is the frequency of the period (Hz), w m is the interference variable between signal m and signal m+1, t m represents the ratio of the travel time from signal m to signal m+1 to the period, q m represents the loop integer variable, δ m A 0 / 1 variable representing the left-turn phase pattern of signal m, where the value is 1 if the left turn precedes the straight move; otherwise, the value is 0.

[0061] Furthermore, in step 3-3, the design objective function is:

[0062]

[0063] Where n is the total number of signals in OD green wave path k, v m and m They represent the predicted traffic volume pcu / h and saturation flow (pcu / h) from signal m to signal m+1 respectively, and p represents the weighted index, with values ​​of 0, 1, 2, and 4;

[0064] Step 3-4 includes the following steps:

[0065] Step 3-4-1: Decision variable solution

[0066] The matrix prepared in step 3-1 is used as the input of the OD green wave path signal coordination model, and Gurobi is used to solve the optimization problem of each green wave band. After solving, the periodic frequency z of the OD green wave path k can be obtained. k , the matrix B composed of each bandwidth k The matrix T consisting of the phase difference k Optimize results;

[0067] Step 3-4-2: Further processing of optimization results

[0068] Since the periodic frequency cannot be directly applied to the actual signal system, it is necessary to convert the common periodic frequency z of the OD green wave path k into k Transformed into practical application cycle C k ;

[0069] Based on the Shapley value Φ(v) of the single-point signal phase in path k obtained in step 2-3, the optimized common period C k The total green light duration is fairly distributed to each phase, and the final green light duration of the signal phase p is obtained Finally, the bandwidth matrix B of each path is transformed into k , phase difference matrix T k , Cycle C k and green light duration matrix Arranged into bandwidth matrix B = [B1, B2, ..., B k ], phase difference matrix T = [T1, T2, ... T k ], periodic matrix C = [C1, C2, ... C k ] and the street light duration matrix G = [G1, G2, ... G k ], as the final result of OD green wave path signal coordination and optimization.

[0070] The present invention also relates to an autonomous coordination and optimization system for unidirectional OD green wave path signals, comprising a computer module, wherein the computer module utilizes the above-mentioned autonomous coordination and optimization method for unidirectional OD green wave path signals.

[0071] Beneficial Effects

[0072] 1. Improve the efficiency and accuracy of green band optimization

[0073] Compared with traditional methods that require human intervention, the present invention realizes a fully autonomous process of signal coordination and optimization, significantly improves the efficiency and accuracy of green band optimization, and reduces errors and delays caused by human factors.

[0074] 2. Enhance the fairness of green light duration allocation

[0075] By introducing cooperative game theory, the present invention fairly allocates the green light duration twice according to the phase capacity, reduces the adverse impact of signal coordination on the passage of vehicles in other directions, and improves the overall efficiency of traffic signal timing.

[0076] 3. Improve the traffic capacity of one-way OD green wave paths

[0077] The present invention designs a signal coordination optimization model specifically for the one-way OD green wave path, which makes up for the deficiency that the traditional two-way coordination model is not applicable to the one-way path, significantly reduces vehicle delays and parking times, and improves the continuity of traffic flow. BRIEF DESCRIPTION OF THE DRAWINGS

[0078] Figure 1It is a schematic diagram of the coordination and optimization process of the present invention. DETAILED DESCRIPTION

[0079] The technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the drawings in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, rather than all of the embodiments. Based on the embodiments in the present invention, all other embodiments obtained by ordinary technicians in this field without making creative work belong to the scope of protection of the present invention. It should be noted that the embodiments in the present invention and the features in the embodiments can be combined with each other without conflict. The present invention is further described below in conjunction with the drawings and specific embodiments, but it is not intended to limit the present invention.

[0080] The whole coordination and optimization process of the present invention, such as Figure 1 shown.

[0081] The present invention mainly comprises three main steps:

[0082] Step 1: Calculation of the initial period of the single-point signal on the OD green wave path

[0083] To support signal coordination and optimization operations in subsequent steps, it is first necessary to prepare relevant data and calculate the initial cycle.

[0084] Step 1-1: Data preparation

[0085] First, according to the predicted OD demand matrix for the next hour, an appropriate user equilibrium traffic allocation model is selected to obtain the corresponding OD travel path. Then, the traffic volume of all lane links allowed to be released by the signal phase is counted, and the corresponding saturation flow is selected according to the lane width (for example, the saturation flow corresponding to a lane width of 3 meters is 1850pcu / h).

[0086] Step 1-2: Determine the initial period of the single-point signal

[0087] The present invention adopts the classic Webster method to calculate the initial period of a single-point signal. Assume that the initial period of a single-point signal m is The calculation formula is:

[0088]

[0089] Where n is the phase number of the single-point signal m, l is the starting loss time (s), z is the distance from the stop line to the conflict point (m), u a is the speed of the vehicle in the import lane (m / s), t s is the vehicle braking time (s), A m is the total duration of the yellow light in a single cycle of signal m (s), q ij andij They represent the traffic volume (pcu / h) and saturation flow (pcu / h) of lane j in the i-th phase of signal m respectively.

[0090] Step 1-3: Unify the signal periods of the OD green wave path

[0091] Green wave optimization requires that all single-point signal cycles in the path are consistent, that is, they share a common cycle. To this end, the present invention uses the maximum value of the signal cycle in the path as the initial common cycle C of each signal in the OD green wave path. k,0 , the calculation formula is:

[0092]

[0093] Where x is the total number of signals in OD green wave path k, It represents the initial period of the xth signal in the OD green wave path k obtained in step 1-2.

[0094] Step 2: Adaptive pre-allocation of single-point signals on OD green wave paths

[0095] In order to improve the fairness and adaptability of the timing results, the present invention regards the phase timing operation as a cooperative game process, and solves it by designing characteristic functions and applying the Shapley valuation method.

[0096] Step 2-1: Calculation of Phase Alliance Key Quantity

[0097] The signal phase in the OD green wave path is regarded as a game participant, forming a cooperative game model G = [P, v]. Among them, P is the phase alliance (the alliance combination is taken from the phase set N = {1, 2, 3, ..., n}), and v is the alliance characteristic function. Key quantity R i It represents the capacity ratio of phase i, and the calculation formula is:

[0098]

[0099] Among them, C cap,i represents the total capacity of all lanes allowed to be released by the single-point signal m in the i-th phase (pcu / h). In addition, to ensure that the allocated phase meets the minimum duration constraint, the starting loss time l of the phase is now combined with the total green light duration T g The ratio of is introduced into the constraint condition of the key quantity. The constraint condition is as follows:

[0100]

[0101] Among them, |S| represents the number of participants in the phase alliance, and |S|≤n, l represents the phase starting loss time (s), usually set to 2 seconds; T gIndicates the initial common period C of the single-point signal m k,0 The total green light duration (s) is calculated as follows: T g =C k,0 -A m -(z / u a -t s ).

[0102] Step 2-2: Calculation of phase alliance characteristic function value

[0103] Using the constraints of key quantities, the alliance characteristic function v(S) of the single-point signal pre-allocation model is designed. The calculation formula is as follows:

[0104]

[0105] This characteristic function follows a voting mechanism: if the sum of the capacity ratios of the alliance S satisfies the key quantity R i If the constraints are met, the alliance is effective; otherwise, its utility is zero.

[0106] Step 2-3: Calculation of the Shapley value of the phase

[0107] First, list all phase alliances of each signal Using the formula in step 2-2, calculate S for each alliance j (j={1,2,…, k})’s characteristic function value v(S j ). Then, calculate the relationship between each phase i∈{1,2,3,…,n} in each phase alliance S j The average marginal contribution value in The calculation formula is:

[0108]

[0109] Where k is the number of phase alliances, n is the total number of signal phases, |S j |Indicates phase alliance S j The number of phases in the j ∪{i}) is the characteristic function value of the alliance containing phase i, v(S j ) is the characteristic function value of the alliance that does not include phase i.

[0110] Finally, the average marginal contribution values ​​of all phases are combined into an n-dimensional Shapley vector It is used to guide the fair allocation of phase green light duration based on phase capacity.

[0111] Step 2-4: Calculation of the initial phase green light duration

[0112] The average marginal contribution value of phase i obtained in step 2-3 is Multiply by the total green light duration T of the initial cycle g , you can get the initial green light duration of this phase The calculation formula is as follows:

[0113]

[0114] Among them, round means rounding operation to ensure that the green light duration is an integer second. However, after the rounding operation, the sum of the initial green light durations of all phases may differ from the original green light duration T g To solve this problem, the difference between the two (can be positive or negative) needs to be added to the phase with the longest green light duration, so as to finally determine the final initial green light duration of the single point signal m phase i The calculation formula is as follows:

[0115]

[0116] If the sum of the initial green light durations of all phases equals the original green light duration T g If the difference is positive, the maximum green light phase duration increases; if the difference is negative, the maximum green light phase duration decreases; if the difference is zero, the maximum green light phase duration remains unchanged.

[0117] Step 3: Coordination and optimization of OD green wave path signals

[0118] In order to realize signal coordination and optimization of unidirectional OD green wave path, the present invention proposes a signal coordination and optimization method based on hybrid linear programming, and its main steps include: data preparation, setting constraint conditions, determining objective function and processing optimization results.

[0119] Step 3-1: Data preparation

[0120] First, along the direction of the OD green wave path k, the input parameters required for the optimization model are calculated in sequence, including the red light duration ratio r of the single point signal m along the green wave direction in the path. m (i.e. the red light duration and the initial cycle C k,0 The ratio of the queue clearing time ratio τ m (That is, the queue clearing time and the initial cycle C k,0 ratio), and the path distance d to the next signal m+1 m , left turn pattern δ for signal m m , the left turn green light duration is longer than L m (i.e. the left turn green light duration and the initial cycle C k,0 ratio), the upper (lower) limit of speed e m (f m ), and the upper (lower) limit of speed change g m (hm ).

[0121] Then, the above data are used in the order of the signal in the OD green wave path k to construct the red light duration ratio matrix R of the path k =[r1,r2,…,r x ], queue clearing time ratio matrix QC k =[τ1,τ2,…,τ x ]、Speed ​​upper and lower limit matrix S speed,k =[[e1,f1],[e2,f2],…,[e x-1 ,f x-1 ]] and the speed change upper and lower limit matrix Δ speed,k =[[g1,h1],[g2,h2],…[g x-1 ,h x-1 ]]. Where x represents the total number of signals in the OD green wave path.

[0122] Step 3-2: Setting constraints

[0123] In order to adapt to the unidirectional OD green wave path signal coordination and optimization, the present invention sets 6 optimization constraints in the model: period constraint, direction interference constraint, integer cycle constraint, green wave speed constraint, green wave speed change constraint, and basic constraints on parameters and decision variables.

[0124] In which, it is assumed that the common period C of OD green wave path k is k,0 The corresponding frequency is z, and the range of the common period is C1≤C k,0 ≤C2, then the period constraint formula is:

[0125]

[0126] Step 3-2-1: Setting of directional interference constraints

[0127] To ensure that the signal coordination results in a green wave effect from the start point to the end point of the OD green wave path, the present invention sets a directional interference constraint. For adjacent signals m and m+1 in the OD green wave path k, the directional constraint is:

[0128]

[0129] Among them, b m is the ratio of the green wave bandwidth between signal m and signal m+1 to the common period, w m is the interference variable between signal m and signal m+1, r m and r m+1 are the red light duration ratios of signal m and signal m+1, w m+1 is the interference variable between signal m+1 and signal m+2, τm+1 Indicates the queue clearing duration ratio of signal m+1.

[0130] Step 3-2-2: Setting integer loop constraints

[0131] In order to ensure that the phase difference between adjacent signals remains unchanged in each cycle, it is necessary to set integer cycle constraints. The one-way integer cycle constraints are:

[0132] w m -w m+1 +t m +δ m L m -δ m+1 L m+1 =(r m+1 -r m )+τ m+1 +q m ,

[0133]

[0134] Among them, t m represents the ratio of the travel time from signal m to signal m+1 to the period, δ m and δ m+1 0 / 1 variables representing the left turn phase mode of signal m and signal m+1 respectively (if the left turn precedes the straight-ahead, the value is 1; otherwise, it is 0), L m and L m+1 They represent the ratio of the left turn green light duration to the cycle of signal m and signal m+1, q m Represents a loop integer variable.

[0135] Step 3-2-3: Setting green wave speed constraints

[0136] In order to ensure that vehicles can reduce the number of stops or even not stop when traveling on the OD green wave path, it is necessary to set a one-way green wave speed constraint as follows:

[0137]

[0138] Among them, d m is the distance (m) between signal m and signal m+1, e m (f m ) is the upper (lower) limit of speed (m / s), and z is the frequency of the cycle (Hz).

[0139] Step 3-2-4: Setting green wave speed change constraints

[0140] Taking into account the positive and negative nature of speed change, the speed change constraint of the one-way green wave is:

[0141]

[0142] Among them, λ m The intermediate variable representing the linearization of the absolute value of the speed change of signal m, t m+1 represents the ratio of the travel time from signal m+1 to signal m+2 to the period, d m+1 is the distance (m) between signal m+1 and signal m+2, g m (b m ) is the upper (lower) limit of the speed change from signal m to signal m+1 (m / s).

[0143] Step 3-2-5: Setting basic constraints on parameters and decision variables

[0144] The parameters and decision variables in the one-way direction from zone O to zone D in the present invention have the following constraints:

[0145]

[0146] Among them, b m is the ratio of the green wave bandwidth between signal m and signal m+1 to the common period, z is the frequency of the period (Hz), w m is the interference variable between signal m and signal m+1, t m represents the ratio of the travel time from signal m to signal m+1 to the period, q m represents the loop integer variable, δ m A 0 / 1 variable representing the left-turn phase pattern of signal m (the value is 1 if a left turn precedes a straight move; the value is 0 otherwise).

[0147] Step 3-3: Design of objective function

[0148] In order to achieve unidirectional OD green wave path signal optimization, the present invention aims to maximize the weighted average value of all bandwidths in the path, and the design objective function is:

[0149]

[0150] Where n is the total number of signals in OD green wave path k, v m and m They represent the predicted traffic volume (pcu / h) and saturated flow (pcu / h) from signal m to signal m+1 respectively, and p represents the weighted index (usually 0, 1, 2, 4).

[0151] Step 3-4: Coordinate optimization and process results

[0152] Step 3-4-1: Decision variable solution

[0153] The matrix prepared in step 3-1 is used as the input of the OD green wave path signal coordination model, and Gurobi is used to solve the optimization problem of each green wave band. After solving, the periodic frequency z of the OD green wave path k can be obtained. k , the matrix B composed of each bandwidth k The matrix T consisting of the phase difference k And other optimization results.

[0154] Step 3-4-2: Further processing of optimization results

[0155] Since the periodic frequency cannot be directly applied to the actual signal system, it is necessary to convert the common periodic frequency z of the OD green wave path k into k Transformed into practical application cycle C k .

[0156] Then, based on the Shapley value Φ(v) of the single-point signal phase in path k obtained in step 2-3, the optimized common period C k The total green light duration is fairly distributed to each phase, and the final green light duration of the signal phase p is obtained

[0157] Finally, the bandwidth matrix B of each path is transformed into k , phase difference matrix T k , Cycle C k and green light duration matrix Arranged into bandwidth matrix B = [B1, B2, ..., B k ], phase difference matrix T = [T1, T2, ... T k ], periodic matrix C = [C1, C2, ... C k ] and the street light duration matrix G = [G1, G2, ... G k ], as the final result of OD green wave path signal coordination and optimization.

[0158] Although the present invention is described herein with reference to specific embodiments, it should be understood that these embodiments are merely examples of the principles and applications of the present invention. It should therefore be understood that many modifications may be made to the exemplary embodiments, and other arrangements may be designed without departing from the spirit and scope of the present invention as defined by the appended claims. It should be understood that the different dependent claims and features described herein may be combined in a manner different from that described in the original claims. It may also be understood that the features described in conjunction with the individual embodiments may be used in other described embodiments.

Claims

1. An autonomous coordination and optimization method for unidirectional OD green wave path signals, characterized in that: The steps include: Step 1: Calculate the initial period of the single-point signal of the OD green wave path; Step 2: When performing adaptive pre-configuration of a single-point signal on an OD green wave path; Step 3: Coordinate and optimize OD green wave path signals.

2. The autonomous coordination and optimization method for unidirectional OD green wave path signals according to claim 1, characterized in that: Step 2 specifically includes the following steps: Step 2-1: Calculate the key quantity of phase alliance; Step 2-2: Calculate the phase alliance characteristic function value; Step 2-3: Calculate the Shapley value of the phase; Step 2-4: Calculate the initial phase green light duration.

3. The autonomous coordination and optimization method for unidirectional OD green wave path signals according to claim 2, characterized in that: In step 2-1, the signal phase in the OD green wave path is regarded as a game participant, forming a cooperative game model G = [P, v]; where P is a phase alliance, the alliance combination is taken from the phase set N = {1, 2, 3, ..., n}, and v is the alliance characteristic function; the key quantity R i It represents the capacity ratio of phase i, and the calculation formula is: Among them, C cap,i It represents the total traffic capacity (pcu / h) of all lanes allowed to be released by the phase i of the single-point signal m. The ratio of the phase start-up loss time to the total green light duration Tg is introduced into the constraint condition of the key quantity; the constraint condition is as follows: Among them, |S| represents the number of participants in the phase alliance, and l represents the phase starting loss time s, which is usually set to 2s; T g Indicates the initial common period C of the single-point signal m k,0 The total green light duration (s) is calculated as follows: T g =C k,0 -A m -(z / u a -t s ).

4. The autonomous coordination and optimization method for unidirectional OD green wave path signals according to claim 2, characterized in that: In step 2-2, the constraint conditions of key quantities are used to design the alliance characteristic function v(S) of the single-point signal pre-allocation model. The calculation formula is as follows: This characteristic function follows a voting mechanism: if the sum of the capacity ratios of the alliance S satisfies the key quantity R i If the constraints are met, the alliance is effective; otherwise, its utility is zero.

5. The autonomous coordination and optimization method for unidirectional OD green wave path signals according to claim 2, characterized in that: In step 2-3, all phase alliances of each signal are listed Using the formula in step 2-2, calculate S for each alliance j The characteristic function value v(S) of (j={1, 2, ..., K}) j );Calculate each phase i∈{1,2,3,...,n} in each phase alliance S j The average marginal contribution value in The calculation formula is: Where k is the number of phase alliances, n is the total number of signal phases, |S j |Indicates phase alliance S j The number of phases in the j ∪{i}) is the characteristic function value of the alliance containing phase i, v(S j ) is the characteristic function value of the alliance that does not contain phase i; The average marginal contribution values ​​of all phases are combined into an n-dimensional Shapley vector It is used to guide the fair allocation of phase green light duration based on phase capacity.

6. The autonomous coordination and optimization method for unidirectional OD green wave path signals according to claim 5, characterized in that: In step 2-4, the average marginal contribution value of phase i obtained in step 2-3 is Multiply by the total green light duration T of the initial cycle g , get the initial green light duration of this phase The calculation formula is as follows: Among them, round means rounding operation to ensure that the green light duration is an integer second; after the rounding operation, the sum of the initial green light durations of all phases may be different from the original green light duration T g If they are not equal, the difference between the two is added to the phase with the longest green light duration, so as to finally determine the final initial green light duration of the single point signal m phase i. The calculation formula is as follows: If the sum of the initial green light durations of all phases equals the original green light duration T g If the difference is positive, the maximum green light phase duration increases; if the difference is negative, the maximum green light phase duration decreases; if the difference is zero, the maximum green light phase duration remains unchanged.

7. The autonomous coordination and optimization method for unidirectional OD green wave path signals according to claim 1, characterized in that: Step 3 specifically includes the following steps: Step 3-1: Prepare data; Step 3-2: Set constraints; Step 3-3: Design objective function; Step 3-4: Process coordination optimization and results.

8. The autonomous coordination and optimization method for unidirectional OD green wave path signals according to claim 7, characterized in that: Step 3-2 includes the following steps: Step 3-2-1: Setting of directional interference constraints For adjacent signals m and m+1 in OD green wave path k, their direction constraints are: Among them, b m is the ratio of the green wave bandwidth between signal m and signal m+1 to the common period, w m is the interference variable between signal m and signal m+1, r m and r m+1 are the red light duration ratios of signal m and signal m+1, W m+1 is the interference variable between signal m+1 and signal m+2, τ m+1 Indicates the queue clearing duration ratio of signal m+1; Step 3-2-2: Setting integer loop constraints The one-way integer cycle constraints are: w m -w m+1 +t m +d m L m -d m+1 L m+1 =(r m+1 -r m )+τ m+1 +q m , Among them, t m represents the ratio of the travel time from signal m to signal m+1 to the period, δ m and δ m+1 0 / 1 variables representing the left turn phase pattern of signal m and signal m+1 respectively. If the left turn precedes the straight-ahead movement, the value is 1; otherwise, it is 0. m and L m+1 They represent the ratio of the left turn green light duration to the cycle of signal m and signal m+1, q m Represents a loop integer variable; Step 3-2-3: Setting green wave speed constraints In order to ensure that vehicles can reduce the number of stops or even not stop when traveling on the OD green wave path, it is necessary to set a one-way green wave speed constraint as follows: Among them, d m is the distance m between signal m and signal m+1, e m (f m ) are the upper and lower limits of the speed (m / s), z is the frequency of the cycle (Hz); Step 3-2-4: Setting green wave speed change constraints Taking into account the positive and negative nature of speed change, the speed change constraint of the one-way green wave is: Among them, λ m The intermediate variable representing the linearization of the absolute value of the speed change of signal m, t m+1 represents the ratio of the travel time from signal m+1 to signal m+2 to the period, d m+1 is the distance (m) between signal m+1 and signal m+2, g m (h m ) are the upper and lower limits of the speed change from signal m to signal m+1 (m / s); Step 3-2-5: Setting basic constraints on parameters and decision variables The parameters and decision variables in the one-way direction from zone O to zone D in the present invention have the following constraints: Among them, b m is the ratio of the green wave bandwidth between signal m and signal m+1 to the common period, z is the frequency of the period (Hz), w m is the interference variable between signal m and signal m+1, t m represents the ratio of the travel time from signal m to signal m+1 to the period, q m represents the loop integer variable, δ m A 0 / 1 variable representing the left-turn phase pattern of signal m, where the value is 1 if the left turn precedes the straight move; otherwise, the value is 0.

9. The autonomous coordination and optimization method for unidirectional OD green wave path signals according to claim 7, characterized in that: In step 3-3, the design objective function is: Where n is the total number of signals in OD green wave path k, v m and m They represent the predicted traffic volume pcu / h and saturation flow (pcu / h) from signal m to signal m+1 respectively, and p represents the weighted index, with values ​​of 0, 1, 2, and 4; Step 3-4 includes the following steps: Step 3-4-1: Decision variable solution The matrix prepared in step 3-1 is used as the input of the OD green wave path signal coordination model, and Gurobi is used to solve the optimization problem of each green wave band. After solving, the periodic frequency z of the OD green wave path k can be obtained. k , the matrix B composed of each bandwidth k The matrix T consisting of the phase difference k Optimize results; Step 3-4-2: Further processing of optimization results Since the periodic frequency cannot be directly applied to the actual signal system, it is necessary to convert the common periodic frequency z of the OD green wave path k into k Transformed into practical application cycle C k ; Based on the Shapley value Φ(v) of the single-point signal phase in path k obtained in step 2-3, the optimized common period C k The total green light duration is fairly distributed to each phase, and the final green light duration of the signal phase p is obtained Finally, the bandwidth matrix B of each path is transformed into k , phase difference matrix T k , Cycle C k and green light duration matrix Arranged into bandwidth matrix B = [B1, B2, ..., B k ], phase difference matrix T = [T1, T2, ...T k ], periodic matrix C = [C1, C2, ...C k ] and the street light duration matrix G = [G1, G2, ...G k ], as the final result of OD green wave path signal coordination and optimization.

10. An autonomous coordination and optimization system for unidirectional OD green wave path signals, characterized in that: It comprises a computer module, which utilizes the autonomous coordination and optimization method for unidirectional OD green wave path signals as described in any one of claims 1 to 9 above.

Images