Trunk line adaptive coordination control method based on dynamic green light starting time interval

Abstract

The invention discloses a trunk line adaptive coordination control method based on a dynamic green light starting time interval. The trunk line adaptive coordination control method comprises the following steps: collecting traffic information; determining influence factors of the green light starting time interval; establishing a green light starting time interval adaptive optimization model; determining a trunk line adaptive coordination timing optimization scheme based on the dynamic green light starting time interval; and outputting a trunk line adaptive coordination timing optimization scheme based on the dynamic green light starting time interval. The green light starting time interval of the adjacent intersection is dynamically optimized according to the dynamically obtained number of queuing vehicles at each intersection, and the green light starting time interval is applied to trunk line coordination control scheme optimization, so that the method has the characteristics of high universality and wide coverage; the dynamic green light starting time interval optimization model established by the invention accords with the change of traffic flow of each intersection of an actual trunk line, the calculation precision is high, and the workload is small; the method can provide an accurate trunk line adaptive coordination control scheme, and is high in scientificity, high in calculation precision and good in reliability.

Description

Technical Field

[0001] The present invention relates to an intelligent transportation control system, and particularly to a main-line adaptive coordination control method based on the dynamic green-light starting time interval. Background Art

[0002] As one of the main management means of urban traffic control systems, main-line coordination control plays an important role in reducing the delay of main-line traffic flow and improving the traffic efficiency of urban roads. At present, the main-line coordination control in urban road networks mainly realizes the optimization of traffic flow by adjusting the green ratio and phase difference.

[0003] Traditional urban main-line coordination control mainly selects the key intersections among several intersections on the main line, and then uses the signal cycle length of the key intersection as the signal cycle of all intersections on the main line. The consequence of this is that the cycle of some non-key intersections on the main line is too long, the delay of the vehicle flow in the direction corresponding to the main line is too large, the traffic flow characteristics cannot be accurately predicted, it is impossible to adapt to the real-time changing traffic flow on the road, which has a certain blindness and also brings certain difficulties to the management of urban traffic management departments. Summary of the Invention

[0004] In order to solve the above problems existing in the prior art, the present invention proposes a main-line adaptive coordination control method based on the dynamic green-light starting time interval, which can accurately predict the traffic flow characteristics and reduce the delay of vehicles.

[0005] In order to achieve the above object, the technical solution of the present invention is as follows:

[0006] A main-line adaptive coordination control method based on the dynamic green-light starting time interval, comprising the following steps:

[0007] A. Collect traffic information

[0008] Select a section with W intersections on the main line as the simulation section, and collect the real-time dynamic traffic information of the simulation section in real time. The traffic information includes the distances between intersections and the real-time traffic flow of each section.

[0009] B. Determine the influencing factors of the green-light starting time interval

[0010] The green-light starting time interval refers to the difference between the starting moments of green lights between adjacent intersections or among multiple intersections, also known as the phase difference; the influencing factors of the green-light starting time interval include the delay time of queuing vehicles at the upstream intersection, the speed change of vehicles, the acceleration duration of vehicles, the distance between adjacent intersections, the number of queuing vehicles at the downstream intersection, the dissipation law of queuing vehicles, and the uniform driving time of vehicles at the upstream intersection. The calculation method is as follows:

[0011] B1. The delay time of the queuing vehicles at the upstream intersection is calculated as follows:

[0012]

[0013] In the formula: i represents the i-th vehicle in the queuing vehicles in front of the intersection; when O = U, ΔT Ui represents the starting delay time of the i-th vehicle in the queuing vehicles in front of the upstream intersection, and Q U is the average number of queuing vehicles per single lane on the through lane in the coordinated direction of the upstream intersection; when O = D, ΔT Di represents the starting delay time of the i-th vehicle in the queuing vehicles in front of the downstream intersection, and Q D is the number of queuing vehicles per single lane on the through lane in the coordinated direction of the downstream intersection; T' is the driver's reaction time; l is the headway between adjacent vehicles, including the length of the leading vehicle and the safety distance; V` is the dissipation wave speed.

[0014] B2. The speed change of the vehicle is calculated as follows:

[0015] Assume that the vehicles behind the b-th vehicle in the queuing queue are regarded as having the same acceleration as the b-th vehicle, and the following formula can be obtained:

[0016]

[0017] In the formula: when O = U, a Ui represents the starting acceleration of the i-th vehicle in the queuing vehicles in front of the upstream intersection; when O = D, a Di represents the starting acceleration of the i-th vehicle in the queuing vehicles in front of the downstream intersection; f(i) is the acceleration change law of the vehicles in front of the b-th vehicle in the queuing vehicles.

[0018] B3. According to the kinematic formula, calculate the acceleration duration of the queuing vehicles as follows:

[0019]

[0020] In the formula: when O = U, T Uai represents the acceleration time of the i-th vehicle in the queuing vehicles in front of the upstream intersection; when O = D, T Dai represents the acceleration duration of the i-th vehicle in the queuing vehicles in front of the downstream intersection.

[0021] B4. The calculation of the queuing dissipation law of the vehicles at the downstream intersection is as follows:

[0022]

[0023] In the formula: H k is the dissipation length of the queuing vehicle fleet from the moment when the green light at the downstream intersection is on to the k moment; is the acceleration of the trailing vehicle of the queuing vehicles at the downstream intersection; TD is the dissipation time of the queuing vehicles before the downstream intersection; is the delayed start time of the last vehicle in the queue at the downstream intersection.

[0024] B5. According to the position of the dissipation point, calculate the uniform driving time of the vehicles at the upstream intersection as follows:

[0025]

[0026] B6. The queuing volume of vehicles at the downstream intersection is based on the data detected at different time periods in actuality.

[0027] C. Establish an adaptive optimization model for the green - light starting offset

[0028] Combined with the vehicle delay at the upstream intersection, the vehicle speed change, the number of queuing vehicles at the downstream intersection, and the dissipation law of queuing vehicles determined in step B, establish an adaptive optimization model for the green - light starting offset for adjacent intersections and multi - intersections on the arterial road respectively.

[0029] D. Determine the arterial - road adaptive coordinated timing optimization scheme based on the dynamic green - light starting offset

[0030] To coordinate the traffic signals of W intersections on the arterial road, the cycle lengths of the W intersections on the arterial road must be equal. First, according to the layout of each intersection, as well as the traffic flow and flow direction, apply Webster's theory to calculate the cycle length required for the traffic signals of each intersection, and then take the maximum cycle length as the cycle length for arterial - road adaptive coordinated control. The intersection with the maximum cycle length is called the key intersection, and take the cycle length of the key intersection as the cycle length of each intersection on the arterial road. Combined with the real - time dynamic traffic flow in different time periods, use Webster's theory to calculate the cycle length, green - signal ratio, and green - light duration of the W intersections on the arterial road, and finally calculate the dynamic green - light starting offset, that is, the dynamic phase difference, between adjacent intersections on the arterial road.

[0031] Since the arterial - road adaptive coordinated control involves W intersections, each intersection needs to optimize its own timing optimization scheme according to the dynamic change of the green - light starting offset between adjacent intersections. Therefore, evaluate the arterial - road adaptive coordinated timing optimization scheme according to the average delay and queue length output by VISSIM software. If the requirements of the minimum average delay and queue length are met, go to step E; otherwise, go to step C.

[0032] E. Output the arterial - road adaptive coordinated timing optimization scheme based on the dynamic green - light starting offset

[0033] Output the arterial - road adaptive coordinated timing optimization scheme based on the adaptive green - light starting offset, providing decision - making information for urban road traffic management personnel.

[0034] Further, the steps of establishing the adaptive optimization model for the green - light start - up time interval in step C are as follows:

[0035] The adaptive optimization model for the green - light start - up time interval includes the adjacent - intersection adaptive green - light start - up time - interval model and the arterial multi - intersection adaptive green - light start - up time - interval model. The specific steps are as follows:

[0036] C1. Establish the adjacent - intersection adaptive green - light start - up time - interval model

[0037] According to the influencing factors of the green - light start - up time interval, after calculating the driving times of each stage for vehicles traveling from the upstream to the downstream, aiming at the goal that the vehicles arriving from the upstream pass through the intersection without stopping, the adjacent - intersection adaptive green - light start - up time - interval model is established as follows:

[0038]

[0039] Constraint conditions:

[0040]

[0041] In the formula: O (U，D) is the green - light start - up time interval between the upstream and downstream intersections; when O = U, ΔT Ui represents the start - up delay time of the i - th vehicle in the queue of vehicles in front of the upstream intersection, T Uai represents the acceleration time of the i - th vehicle in the queue of vehicles in front of the upstream intersection, T Uci represents the constant - speed driving time of the i - th vehicle in the queue of vehicles in front of the upstream intersection, Q U is the average number of queued vehicles in a single lane on the straight - through lane in the coordinated direction of the upstream intersection; when O = D, ΔT Di represents the start - up delay time of the i - th vehicle in the queue of vehicles in front of the downstream intersection, T Dai represents the acceleration time of the i - th vehicle in the queue of vehicles in front of the downstream intersection, Q D is the number of queued vehicles in a single lane on the straight - through lane in the coordinated direction of the downstream intersection; H p is the distance between the dissipation point and the end of the queue of vehicles in front of the downstream intersection; l is the head - to - head distance between adjacent vehicles; l c is the vehicle length; T D is the dissipation time of the queue of vehicles in front of the downstream intersection; L D is the length of the single - lane queue at the downstream intersection; H U，D is the distance between the stop lines at the entrances of the upstream and downstream intersections.

[0042] C2. Establish the arterial multi - intersection adaptive green - light start - up time - interval model

[0043] Suppose there are \(W\) intersections on the main line, and the intersections are numbered 1, 2, 3, …, \(W\) in the downstream direction. It is defined that the dissipation is completed when the last vehicle in the queue at the downstream intersection accelerates to the same speed as the first vehicle of the vehicle arriving at the upstream intersection, that is: the queuing vehicles in front of each intersection merge into the coordinated vehicle fleet to form a new coordinated vehicle fleet. Therefore, when calculating the green start time interval between the \(n\)th intersection and the \((n - 1)\)th intersection, only the number of queuing vehicles in front of the \((n - 1)\)th intersection, the vehicle speed change, and the driving distance need to be considered, and it has nothing to do with the intersections before the \((n - 1)\)th intersection, where \(n = 1, 2, 3, …, W\). Then the multi - intersection adaptive green start time interval model on the main line is regarded as a set of multi - group adjacent intersection adaptive green start time interval models, as shown in the following formula:

[0044]

[0045] Constraint conditions:

[0046]

[0047] In the formula: \(O\) (n-1，n) is the green start time interval between the \((n - 1)\)th intersection and the \(n\)th intersection, where \(n = 1, 2, 3, …, w\); \(\Delta T\) (n-1)i is the start delay time of the \(i\)th vehicle in the queuing vehicles in front of the \((n - 1)\)th intersection; \(T\) (n-1)ai is the acceleration time of the \(i\)th vehicle in the queuing vehicles in front of the \((n - 1)\)th intersection; \(V\) nf is the vehicle uniform driving speed on the road between the \((n - 1)\)th intersection and the \(n\)th intersection; \(a\) (n-1)i is the acceleration of the \(i\)th vehicle in the queuing vehicles in front of the \((n - 1)\)th intersection; \(T\) (n-1)ci is the uniform driving time of the \(i\)th vehicle in the queuing vehicles in front of the \((n - 1)\)th intersection; \(H\) n-1，n is the distance between the \((n - 1)\)th intersection and the \(n\)th intersection; \(H\) np is the distance between the dissipation point in front of the \(n\)th intersection and the last vehicle in the queue; \(T\) n is the dissipation time of the queuing vehicles in front of the \(n\)th intersection; \(Q\) n is the number of queuing vehicles in front of the \(n\)th intersection; \(L\) n is the queuing length in front of the \(n\)th intersection.

[0048] Furthermore, the method for determining the main - line adaptive coordinated timing optimization scheme based on the dynamic green start time interval in step D is as follows:

[0049] In the process of arterial adaptive coordinated control, for the g-th cycle, the cycle length of this cycle is determined by the cycle lengths of the key intersections in the (g - 1)-th cycle. Obviously, when calculating the cycle length of the g-th cycle using the traffic flow data obtained in the (g - 1)-th cycle, it is necessary to ensure that before obtaining the traffic flow of the (g - 1)-th cycle, the signal timing of the starting intersection in the arterial coordinated control is still implementing the timing plan of the (g - 1)-th cycle, that is, the signal timing of the starting intersection has not entered the g-th cycle.

[0050] After obtaining the road traffic flow data of the key intersections in the (g - 1)-th cycle of the arterial adaptive coordinated control, calculate the optimal cycle length required for the key intersections as the cycle length of the g-th cycle:

[0051]

[0052] In the formula: C g is the cycle length of the g-th cycle; L is the total lost time of the key intersections in the (g - 1)-th cycle; Y is the sum of the maximum traffic flow ratios of each phase of the key intersections in the (g - 1)-th cycle.

[0053] Among them, the total lost time and the sum of the maximum traffic flow ratios of each intersection phase are:

[0054]

[0055] In the formula: l′ is the start-up lost time; A is the yellow light time; I is the green light interval time, that is, the time from the end of the green light of the previous phase to the start of the green light of the next phase; u is the number of phases in the (g - 1)-th cycle;

[0056]

[0057] In the formula: y u is the traffic flow ratio of the critical lane of the u-th phase of the key intersection in the (g - 1)-th cycle.

[0058] In the process of arterial coordinated control, the allocation of the green light time of each phase of each intersection is also based on the principle of minimizing vehicle delay. Therefore, the effective green light duration of each phase should be allocated according to the traffic flow ratio of each phase. For any n-th intersection, according to the traffic flow Q of different directions of each approach lane measured and collected in the (g - 1)-th cycle xy After that, calculate the traffic flow ratio y of the critical lane of the u-th phase within the (g - 1)-th cycle nu . The effective green light time G of the u-th phase of the n-th intersection in the g-th cycle enu The calculation formula is as follows:

[0059]

[0060] In the formula: G enu is the effective green light time of the u-th phase of the n-th intersection in the g-th cycle; ynu is the traffic flow ratio of the critical lane of the u-th phase in the (g - 1)-th cycle of the n-th intersection; C g is the cycle length of the g-th cycle; L n is the total lost time of the n-th intersection; Y n is the sum of the maximum traffic flow ratios of all phases in the (g - 1)-th cycle of the n-th intersection.

[0061] The green signal ratio of each phase in the g-th cycle of the n-th intersection is:

[0062]

[0063] In the formula: λ nu is the green signal ratio of the u-th phase in the g-th cycle of the n-th intersection.

[0064] The green light display time of each phase is:

[0065] G nu = G enu - A u + l u '

[0066] In the formula: G nu is the green light display time of the u-th phase in the g-th cycle of the n-th intersection; A u is the yellow light time of the u-th phase; l′ u is the start-up lost time of the u-th phase.

[0067] Compared with the prior art, the present invention has the following beneficial effects:

[0068] 1. Since the present invention is based on a large amount of vehicle passing information at intersections, through screening and processing, the starting, accelerating, and queue dissipation laws of vehicles are summarized. According to the dynamically obtained number of queuing vehicles at each intersection, the starting time intervals of green lights at adjacent intersections are dynamically optimized and applied to the optimization of the arterial coordination control scheme. This is different from the existing arterial coordination control and has the characteristics of strong generality and wide coverage;

[0069] 2. The dynamic green light starting time interval optimization model established by the present invention conforms to the changes in traffic flow at each intersection of the actual arterial, has high calculation accuracy and small workload;

[0070] 3. Since the present invention can provide an accurate arterial adaptive coordination control scheme, it has strong scientificity, high calculation accuracy, and good reliability;

[0071] 4. Since the present invention adaptively coordinates and optimizes the signal timing of each intersection of the arterial based on the dynamic green light starting time interval, it can effectively reduce the vehicle delay and carbon emission level of the urban road network and has good economy.

[0072] 5. By adopting the arterial adaptive coordinated control method proposed by the present invention, the urban traffic management department can conduct dynamic macro management on the vehicles within the urban road network. BRIEF DESCRIPTION OF THE DRAWINGS

[0073] The present invention has a total of Figure 9 sheets, among which:

[0074] Figure 1 is the technical roadmap of the arterial adaptive coordinated control based on the dynamic green light start-up time interval.

[0075] Figure 2 is the plan view of the simulation section.

[0076] Figure 3 is the schematic diagram of the simulation section.

[0077] Figure 4 is the schematic diagram of the vehicle speed change process at adjacent intersections.

[0078] Figure 5 is the current situation and optimized phase timing diagram of the Torch Road - Huangpu Road intersection.

[0079] Figure 6 is the current situation and optimized phase timing diagram of the Wanda Plaza - Huangpu Road intersection.

[0080] Figure 7 is the current situation and optimized phase timing diagram of the Jinhui Shopping Mall - Huangpu Road intersection.

[0081] Figure 8 is the current situation and optimized phase timing diagram of the Lingshui Road - Huangpu Road intersection.

[0082] Figure 9 is the comparison diagram of the optimization degree of the two-way road section model. DETAILED DESCRIPTION OF THE INVENTION

[0083] The present invention will be further described below with reference to the accompanying drawings. As Figure 1 shown, an arterial adaptive coordinated control method based on the dynamic green light start-up time interval includes the following steps:

[0084] (1) Traffic information collection

[0085] The present invention selects Figure 2 and Figure 3 four intersections along Huangpu Road in Dalian shown as the simulation section. The upstream direction is from the Torch Road intersection to the Lingshui Road intersection (west - east), and the downstream direction is from the Lingshui Road intersection to the Torch Road intersection (east - west). The traffic information such as the distance, green ratio, and real-time traffic flow of each intersection on the section is obtained in real time by using the video detection method, as shown in Table 1.

[0086] Table 1 Parameter Information of Each Road Intersection

[0087]

[0088]

[0089] (2) Analysis of Influencing Factors of Green - light Starting Time Interval

[0090] The green - light starting time interval is a main parameter affecting the arterial control system, and it is affected by multiple factors. The present invention divides the process of a vehicle driving from an upstream intersection to a downstream intersection into three stages.

[0091] The first stage mainly refers to the stage from when the green light at the upstream intersection turns on to when the vehicle starts. The idle - delay time of the vehicle in the first stage includes the driver's reaction time and the time to wait for the vehicle in front to start. The driver's reaction time is usually an empirical value of 1 - 2 seconds. The present invention needs to analyze the starting delay of vehicles in different queuing positions, which should be included in the green - light starting time interval between two adjacent intersections.

[0092] The second stage refers to the stage when the vehicle starts to accelerate until it reaches a uniform speed. The present invention takes into account that the accelerations of vehicles in different queuing positions are also different, and it is necessary to predict the acceleration of each position vehicle according to the specific road information.

[0093] The third stage refers to the stage of uniform - speed driving after the vehicle has completed acceleration. The present invention combines the number of queuing vehicles in front of the downstream intersection and the vehicle dissipation law, calculates the dissipation time and dissipation distance of the queuing vehicles after the green light turns on, and further obtains the time difference between when the vehicle fleet starts to drive at a uniform speed and when the green light at the downstream intersection turns on.

[0094] For the queuing vehicles at the upstream intersection, accurately obtaining the starting delay time of the vehicle fleet, the acceleration time, and the time difference between when it drives at a uniform speed until the green light at the downstream intersection turns on, the sum of the three can be used as the dynamic green - light starting time of the arterial coordinated control system.

[0095] (3) Establishing an Adaptive Optimization Model for Green - light Starting Time Interval

[0096] 1) Model Assumptions

[0097] The present invention aims to discuss the influence of vehicle delay at the upstream intersection, vehicle speed change, the number of queuing vehicles at the downstream intersection, and the queuing vehicle dissipation law on the green - light starting time interval by constructing a mathematical model. Since the actual traffic flow has randomness and discreteness, in order to ensure the stable operation state of vehicles, the following assumptions are made in the present invention:

[0098] ① The road traffic flow only discusses the homogeneous traffic flow composed of the same type of vehicles, so the vehicle length is uniformly l c ;

[0099] ②After the vehicle accelerates through the intersection, it is regarded as driving at a constant speed at the same speed.

[0100] ③Pedestrians and non-motor vehicles comply with the traffic lights and do not affect the operation of motor vehicles.

[0101] 2) Establish an adaptive green light starting time interval model for adjacent intersections

[0102] ①Startup delay analysis

[0103] The present invention mainly considers the green light startup delay of the queuing vehicle fleet at the upstream intersection, and calculates the green light startup delay of the queuing vehicle fleet based on the traffic wave theory. The traffic wave describes the conversion process of two traffic states. When the stationary vehicle fleet queuing in front of the upstream intersection starts, the interface where the traffic flow changes from the high-density state to the low-density state moves towards the end of the queue, and at this time, a dissipation wave is generated. The dissipation wave speed is:

[0104]

[0105] In the formula: V′ is the dissipation wave speed, with the unit of km / h; V1 is the speed of the released traffic flow; ρ1 is the density of the released traffic flow, with the unit of veh / km; ρ0 is the vehicle density during congestion queuing; V f is the normal uniform driving speed of vehicles on the road between the upstream and downstream intersections.

[0106] When the green light at the upstream intersection starts to shine, the driver first needs a reaction time, generally taking the empirical value T′. Due to the continuity of the traffic wave and the additivity of the startup delay time, the reaction time of the leading vehicle can be directly used as the delayed startup time of the leading vehicle, and the reaction time of the subsequent vehicles is included in the startup delay time.

[0107] After obtaining the wave speed of the dissipation wave of the queuing vehicles at the upstream intersection, combined with the queuing distance of each queuing vehicle from the stop line of the intersection, based on the traffic wave theory and the Webster delay model for the startup delay time model of the waiting queue in the non-saturated state, the startup delay time ΔT of the i-th vehicle can be obtained i as:

[0108]

[0109] In the formula: i is the i-th vehicle in the queuing vehicles in front of the intersection; when O = U, ΔT Ui represents the startup delay time of the i-th vehicle in the queuing vehicles in front of the upstream intersection, Q U is the average number of queuing vehicles in a single lane on the straight lane in the coordinated direction of the upstream intersection; when O = D, ΔT Di represents the startup delay time of the i-th vehicle in the queuing vehicles in front of the downstream intersection, Q D$n$ is the number of single-lane queuing vehicles on the through lane in the coordinated direction of the downstream intersection; $T'$ is the driver's reaction time; $l$ is the headway between adjacent vehicles, including the length of the leading vehicle and the safety distance.

[0110] In the present invention, it is considered that vehicles will maintain the minimum stopping distance when queuing and only homogeneous traffic flow is discussed. Therefore, the distance between adjacent vehicles in the present invention is uniformly $l$.

[0111] ② Analysis of acceleration duration

[0112] When the queuing vehicles at the upstream intersection start to accelerate, the accelerations of vehicles in different parking positions are different. Since there is no vehicle in front of the vehicle at position 1 when it starts, the driver is accustomed to quickly increasing the speed to compensate for the time loss of waiting for the traffic light. In order to maintain the vehicle distance, the acceleration of the vehicle at a later position is less than that of the vehicle in front, and the acceleration magnitude and change rate of the vehicle at a later position are getting closer and closer. Therefore, in the present invention, the vehicles behind the $b$-th vehicle in the queuing queue are regarded as having the same acceleration as the $b$-th vehicle, and the following formula (4) can be obtained:

[0113]

[0114] In the formula: when $O = U$, $a$ Ui represents the starting acceleration of the $i$-th vehicle in the queuing vehicles in front of the upstream intersection; when $O = D$, $a$ Di represents the starting acceleration of the $i$-th vehicle in the queuing vehicles in front of the downstream intersection; $f(i)$ is the acceleration change law of the vehicles in front of the $b$-th vehicle in the queuing vehicles.

[0115] According to the kinematic formula, the calculation formula for the acceleration duration of the queuing vehicles is as follows:

[0116]

[0117] In the formula: when $O = U$, $T$ Uai represents the acceleration time of the $i$-th vehicle in the queuing vehicles in front of the upstream intersection; when $O = D$, $T$ Dai represents the acceleration time of the $i$-th vehicle in the queuing vehicles in front of the downstream intersection.

[0118] ③ Analysis of constant-speed time

[0119] The present invention proposes a new method for judging the dissipation time, that is, a dissipation point is established in the section, and this dissipation point satisfies that when the vehicles at the upstream intersection reach the end of the queuing vehicles at the downstream intersection, the last vehicle in the queuing at the downstream intersection has accelerated to the normal constant-speed driving speed $V$ f of the section, and at this time, the vehicle fleet arriving at the upstream intersection can pass through the downstream intersection without decelerating.

[0120] Assume that when the green light at the downstream intersection turns on, the leading vehicle of the vehicle fleet passing through the upstream intersection reaches a certain position. This position is denoted as the dissipation point, that is, when the leading vehicle of the vehicle fleet passing through the upstream intersection reaches the dissipation point of the downstream intersection, the green light at the downstream intersection turns on. The distance between the dissipation point and the rear of the queue of the last vehicle at the downstream intersection is H p 。

[0121] The process of the vehicle speed change at adjacent intersections is as Figure 4 shown:

[0122] The dissipation time of the queuing vehicles in front of the downstream intersection refers to the longest time required for all queuing vehicles to accelerate to V f :

[0123] T D = max(ΔT Di + T Dai ), i = 1, 2, 3,..., Q D (6)

[0124] In the formula: T D is the dissipation time of the queuing vehicles in front of the downstream intersection; ΔT Di is the start-up delay time of the i-th vehicle in the queuing vehicles at the downstream intersection; T Dai is the starting acceleration time of the i-th vehicle in the queuing vehicles at the downstream intersection.

[0125] In order to dynamically represent the dissipation law, it is necessary to consider the queuing length dissipated at each moment after the dissipation starts. Assume that T k is the time difference from the moment when the green light at the downstream intersection turns on to the k-th moment. Then the dynamic change law of the dissipation length can be expressed as:

[0126]

[0127] In the formula: H k is the queuing fleet dissipation length from the moment when the green light at the downstream intersection turns on to the k-th moment; is the acceleration of the last vehicle in the queuing vehicles at the downstream intersection; is the delayed start time of the last vehicle in the queuing vehicles at the downstream intersection.

[0128] To maximize the traffic efficiency at the downstream intersection, it is required that when the queuing vehicles at the downstream intersection are completely dissipated, the leading vehicle at the upstream intersection reaches the rear of the queue of the vehicles at the downstream intersection. At this time, T k = T D . According to the number of queuing vehicles in front of the downstream intersection and the dissipation time of the queuing vehicles in front of the downstream intersection, a dynamic dissipation point can be determined. The distance between this dissipation point and the last vehicle in the queue at the downstream intersection is H p :

[0129] When T k = TD When

[0130] H p +H k -(l - l c ) = V f T k (8)

[0131] That is:

[0132]

[0133] In the formula: H p is the distance between the dissipation point and the last vehicle in the queue at the downstream intersection; l c is the vehicle length.

[0134] After obtaining the position of the dissipation point, the uniform - speed driving distance and time of the vehicles at the upstream intersection can be calculated:

[0135]

[0136] In the formula: T Uci is the uniform - speed driving time of the i - th vehicle in the queue of vehicles at the upstream intersection; H U，D is the distance between the stop lines of the approach lanes at the upstream and downstream intersections; L D is the queuing length of a single lane at the downstream intersection.

[0137] ④ Establish a dynamic green - light start - up time - interval model for adjacent intersections

[0138] Based on the analysis of the influence of the above - mentioned various factors on the green - light start - up time - interval, after obtaining the driving times of each stage when the vehicle travels from the upstream to the downstream, with the goal that the vehicles arriving at the upstream do not stop when passing through the intersection, a dynamic green - light start - up time - interval model for adjacent intersections is established as follows:

[0139]

[0140] Constraint conditions:

[0141]

[0142] In the formula: O (U，D) is the green - light start - up time - interval between the upstream and downstream intersections; when O = U, ΔT Ui represents the start - up delay time of the i - th vehicle in the queue of vehicles in front of the upstream intersection, T Uai represents the acceleration time of the i - th vehicle in the queue of vehicles in front of the upstream intersection, T Uci represents the uniform - speed driving time of the i - th vehicle in the queue of vehicles in front of the upstream intersection, Q U is the average number of queuing vehicles in a single lane on the straight - through lane in the coordinated direction of the upstream intersection; when O = D, ΔT DiDenote the start-up delay time of the $i$-th vehicle in the queue of vehicles before the downstream intersection as $T$. Dai Denote the acceleration time of the $i$-th vehicle in the queue of vehicles before the downstream intersection as $Q$. D $H$ is the number of vehicles queuing in a single lane on the straight-ahead lane in the coordinated direction of the downstream intersection. p $l$ is the distance between the dissipation point and the last vehicle in the queue at the downstream intersection; $l$ is the headway between adjacent vehicles. c $L$ is the vehicle length; $T$ D Denote the dissipation time of the queue of vehicles before the downstream intersection as $L$. D $H$ is the length of the queue in a single lane at the downstream intersection. U，D Denote the distance between the stop lines at the approaches of the upstream and downstream intersections as $L$.

[0143] When the signal lights in two directions of a main-line coordination at an intersection are in the same phase, the start-up time intervals of the green lights in the two directions are correlated with each other. When the start-up time interval of the green light in a certain direction is determined, the time difference in the other direction is also automatically determined. The start-up time intervals of the green lights in the two directions of a two-way road should satisfy the following relationship:

[0144] $O$ (U,D) $+O$ (D,U) $=n\cdot C(14)$

[0145] In the formula: $O$ (U，D) and $O$ (D，U) Denote the start-up time intervals of the green lights in two opposite running directions on a two-way section; $C$ is the cycle length; $n$ is an integer, including positive integers, negative integers or zero.

[0146] ⑤ Establish the main-line multi-intersection adaptive green-light start-up time interval model

[0147] Suppose there are $W$ intersections on the main line, and the intersections are numbered 1, 2, 3, …, $w$ in the downstream direction. Since a new definition of actual dissipation is proposed in the present invention, when the last vehicle in the queue at the downstream intersection accelerates to the same speed as the first vehicle of the vehicle arriving at the upstream intersection, it is regarded as the completion of dissipation. That is to say, the queue of vehicles in front of each intersection will merge into the coordinated vehicle platoon to form a new coordinated vehicle platoon. Therefore, when calculating the start-up time interval of the green light between the $n$-th intersection and the $(n - 1)$-th intersection, only the number of vehicles queuing in front of the $(n - 1)$-th intersection, the vehicle speed change and the driving distance need to be considered, and it has nothing to do with the intersections before the $(n - 1)$-th intersection, $n = 1, 2, 3, …, w$. Then the main-line multi-intersection adaptive green-light start-up time interval model can be regarded as a set of adjacent intersection adaptive green-light start-up time interval models, as shown in Equations (15) and (16):

[0148]

[0149] Constraint conditions:

[0150]

[0151] In the formula: O (n-1，n) is the green - light starting time interval between the (n - 1)th intersection and the nth intersection, where n = 1, 2, 3, … w; ΔT (n-1)i is the starting delay time of the ith vehicle in the queue in front of the (n - 1)th intersection; T (n-1)ai is the acceleration time of the ith vehicle in the queue in front of the (n - 1)th intersection; V nf is the constant - speed driving speed of the vehicle on the road between the (n - 1)th intersection and the nth intersection; a (n-1)i is the acceleration of the ith vehicle in the queue in front of the (n - 1)th intersection; T (n-1)ci is the constant - speed driving time of the ith vehicle in the queue in front of the (n - 1)th intersection; H n-1，n is the distance between the (n - 1)th intersection and the nth intersection; H np is the distance between the dissipation point in front of the nth intersection and the end vehicle of the queue; T n is the dissipation time of the queue of vehicles in front of the nth intersection; Q n is the number of vehicles in the queue in front of the nth intersection; L n is the queue length in front of the nth intersection.

[0152] (4) Optimized trunk - line coordinated timing scheme based on adaptive green - light starting time interval

[0153] Suppose there are W intersections on the trunk - line, and the intersections are numbered 1, 2, 3, …, w in the downstream direction. For the nth intersection in the coordinated direction, due to the dynamic changes of O(n - 1, n) and O(n, n + 1) in each cycle, the traffic flow at this intersection changes in real - time in each cycle. When the present invention conducts trunk - line coordinated control based on the dynamic green - light starting time - interval model, it is also necessary to consider the influence of the traffic - flow dynamics on the trunk - line coordinated control scheme.

[0154] 1) Phase design and channelization scheme optimization

[0155] In intersection signal control, when conducting signal - phase design and channelization - scheme design, it should be considered according to the specific road conditions, traffic conditions, actual traffic management measures and traffic requirements at the intersection. Due to the characteristics of mixed traffic at most road intersections, the combination of vehicle flows at the intersection is relatively complex, and it is necessary to determine the phase and channelization scheme according to the geometric shape of each approach, the size and direction of the traffic flow at the intersection.

[0156] 2) Optimization of basic parameters of trunk - line coordination

[0157] ① Traffic flow

[0158] The present invention contemplates installing video detectors beside the signal lights at each approach lane of each intersection on the arterial road, so as to obtain dynamic traffic flow data in real time. Considering that the change of traffic flow data in the road network is a stationary random process, and the traffic flow data at the same road intersection generally does not change suddenly in adjacent cycles, for the nth intersection in the coordinated section, the traffic flow passing through in the (g - 1)th cycle can be used as the basis for calculating the traffic flow required for signal timing in the gth cycle.

[0159] Label the approach lanes for main road vehicles entering the intersection as m1 and m2, and the approach lanes for branch road vehicles entering the intersection as s1 and s2. Using the video detector, the flow data Q of different directions of each approach lane in the (g - 1)th cycle of the nth intersection can be obtained xy , where x represents the approach lane, x = m1, m2, s1, s2; y represents the traffic flow in different directions of a certain approach lane, and y = 1, 2, 3 represent the left-turn traffic flow, straight-through traffic flow, and right-turn traffic flow respectively.

[0160] Then, for the nth intersection, the hourly traffic flow required for calculating the signal timing in the gth cycle is:

[0161]

[0162] In the formula: Q g is the hourly traffic flow of the nth intersection calculated based on the traffic flow in the (g - 1)th cycle; C g-1 is the cycle length of the (g - 1)th cycle in the arterial coordinated control; Q xy is the traffic flow of different directions of each approach lane in the (g - 1)th cycle in the arterial coordinated control.

[0163] ② System cycle length

[0164] In the present invention, for the gth cycle, the cycle length of this cycle is determined by the cycle length of the key intersection in the (g - 1)th cycle. Obviously, using the traffic flow data obtained in the (g - 1)th cycle to calculate the cycle length of the gth cycle requires ensuring that before obtaining the traffic flow data of the (g - 1)th cycle, the signal timing at the starting intersection is still executing the timing scheme of the (g - 1)th cycle, that is, the signal timing at the starting intersection has not entered the gth cycle.

[0165] Assume that m is a certain intersection, and the nth intersection is the key intersection. For

[0166]

[0167] In the formula: Equation (18) ensures that before entering the gth cycle, the traffic flow data of the key intersection in the (g - 1)th cycle can be collected completely. ΔT miis the starting delay time of the $i$-th vehicle in the queue of vehicles before the $m$-th intersection; $T$ mai is the acceleration time of the $i$-th vehicle in the queue of vehicles before the $m$-th intersection; $V$ (m+1)f is the uniform driving speed of vehicles on the road between the $m$-th intersection and the $(m + 1)$-th intersection; $V$ nf is the uniform driving speed of vehicles on the road between the $(n - 1)$-th intersection and the $n$-th intersection; $a$ mi is the acceleration of the $i$-th vehicle in the queue of vehicles before the $m$-th intersection; $H$ m，m+1 is the distance between the $m$-th intersection and the $(m + 1)$-th intersection; $H$ (m+1)p is the distance between the dissipation point in front of the $(m + 1)$-th intersection and the end vehicle of the queue; $L$ m+1 is the queue length in front of the $(m + 1)$-th intersection; $H$ n is the intersection distance of the $n$-th intersection, that is, the distance between the stop line of the entrance lane and the stop line of the exit lane; $C$ g-1 is the cycle length of the $(g - 1)$-th cycle.

[0168] After obtaining the road information and traffic flow data of the key intersections in the $(g - 1)$-th cycle, calculate the optimal cycle length required for the key intersections as the cycle length of the $g$-th cycle:

[0169]

[0170] In the formula: $C$ g is the cycle length of the $g$-th cycle; $L$ is the total lost time of the key intersections in the $(g - 1)$-th cycle; $Y$ is the sum of the maximum traffic flow ratios of each phase of the key intersections in the $(g - 1)$-th cycle.

[0171] Among them, the total lost time and the sum of the maximum traffic flow ratios of each intersection phase are:

[0172]

[0173] In the formula: $l'$ is the starting lost time; $A$ is the yellow light time; $I$ is the green light interval time (from the end of the green light of the previous phase to the start of the green light of the next phase); $u$ is the number of phases in the $(g - 1)$-th cycle;

[0174]

[0175] In the formula: $y$ u is the traffic flow ratio of the critical lane of the $u$-th phase of the key intersections in the $(g - 1)$-th cycle.

[0176] ③ The green light duration of each phase of the intersection

[0177] In arterial coordinated control, the allocation of the green light time for each phase at each intersection is also based on the principle of minimizing vehicle delay. Therefore, the effective green light duration for each phase should be allocated according to the traffic flow ratio of each phase. For any nth intersection, based on the traffic flow Q in different directions of each approach lane obtained by the video detector during the (g - 1)th cycle xy After that, the traffic flow ratio y of the critical lane of the u-th phase during the (g - 1)th cycle can be calculated nu . The effective green light time G of the u-th phase during the g-th cycle at the nth intersection eun The calculation method is as follows:

[0178]

[0179] In the formula: G enu is the effective green light time of the u-th phase during the g-th cycle at the nth intersection; y nu is the traffic flow ratio of the critical lane of the u-th phase during the (g - 1)th cycle at the nth intersection; C g is the cycle length of the g-th cycle; L n is the total lost time at the nth intersection; Y n is the sum of the maximum traffic flow ratios of each phase during the (g - 1)th cycle at the nth intersection.

[0180] The green signal ratio of each phase during the g-th cycle at the nth intersection is:

[0181]

[0182] In the formula: λ nu is the green signal ratio of the u-th phase during the g-th cycle at the nth intersection.

[0183] The displayed green light time for each phase is:

[0184] G nu = G enu - A u + l u ' (24)

[0185] In the formula: G nu is the displayed green light time of the u-th phase during the g-th cycle at the nth intersection; A u is the yellow light time of the u-th phase; l′ u is the start-up lost time of the u-th phase.

[0186] The current situation of the four intersections along the Huangpu Road arterial line to which the present invention is applied and the results of the optimized timing plan are shown in Table 2 and Figures 5 - 8 as follows:

[0187] Table 2 Current situation of the four intersections along the Huangpu Road arterial line and the optimized timing plan

[0188]

[0189]

[0190] (5) Output the arterial adaptive coordinated timing plan based on the dynamic green - light start time interval

[0191] Output the optimized arterial coordinated timing plan based on the dynamic green - light start time interval, providing decision - making information for traffic management personnel.

[0192] To verify the availability and effectiveness of the dynamic green - light start time interval model established in the present invention, taking the intersection queue length and average delay as indicators, the delay time can provide information on aspects such as traffic flow, congestion level, and traffic management efficiency; the queue length directly reflects the traffic congestion level and signal control effect.

[0193] 1) Analysis of experimental results

[0194] In VISSIM, the current situation and the model plan of four intersections along the Huangpu Road arterial were respectively simulated for 1 hour. The evaluation indicators under the two different control plans are shown in Table 3, Table 4 and Figure 9 as shown.

[0195] Table 3 Comparison of indicators before and after optimization of each intersection

[0196]

[0197] Table 4 Comparison of indicators before and after optimization of two - way sections

[0198]

[0199]

[0200] According to the simulation results, whether it is the evaluation indicators of each intersection or the evaluation indicators of two - way sections, the model plan proposed by the present invention is superior to the current situation plan. In the west - east direction, the queue length on the arterial road under the optimized plan decreased by 5.83%, and the average delay decreased by 10.27%; in the east - west direction, the queue length on the arterial road under the optimized plan decreased by 16.83%, and the average delay decreased by 15.52%. It can be seen that the overall green - wave control effect of the section has been improved, indicating that the dynamic green - light start time interval model proposed by the present invention has a significant improvement in the control effect of arterial coordination and is helpful in alleviating traffic congestion problems and shortening travel time.

[0201] In summary, the present invention discusses the influence of vehicle delay at upstream intersections, vehicle speed changes, the number of queuing vehicles at downstream intersections, and the dissipation law of queuing vehicles on the green start-up time interval by constructing a mathematical model, establishes a dynamic green start-up time interval optimization model, and finally applies Webster's theory based on this model to solve the optimized scheme for arterial adaptive coordinated signal timing. It has the characteristics of wide coverage and strong applicability. Different from the static coordinated control of arterial intersections, the present invention designs and develops an arterial adaptive coordinated signal timing scheme based on the dynamic green start-up time interval, with less workload, good economy, and strong generality. The present invention only needs to utilize the traffic information such as the distance, green ratio, and number of vehicles at each intersection of the road section obtained in real time, calculate the green start-up time interval of different cycles on each road section through the dynamic green start-up time interval model, adjust the adaptive signal timing scheme of each intersection of the arterial according to the dynamic green start-up time interval, and evaluate the control benefit of two-way arterial coordination, so as to realize the real-time dynamic macroscopic management of urban road network vehicles.

[0202] The present invention is not limited to this embodiment, and any equivalent conceptions or changes within the technical scope disclosed in the present invention are included in the protection scope of the present invention.

Claims

1. A trunk line adaptive coordinated control method based on dynamic green light start time interval, characterized in that: The following steps are involved: A. Collecting traffic information A road section with W intersections on the trunk line is selected as a simulation section, and real-time dynamic traffic information of the simulation section is collected in real time, wherein the traffic information includes the distance of each intersection and the real-time traffic flow of each section; B. Factors affecting the green light start time The green light start time refers to the difference between the green light start time of two adjacent intersections or multiple intersections, also known as the phase difference; the factors affecting the green light start time include the delay time of vehicles queuing at the upstream intersection, the speed change of the vehicle, the acceleration time of the vehicle, the distance between adjacent intersections, the number of vehicles queuing at the downstream intersection and the dissipation law of the queuing vehicles, and the uniform speed driving time of vehicles at the upstream intersection. The calculation method is as follows: B1. The delay time of vehicles queuing at the upstream intersection is calculated as follows: Where: i is the i-th vehicle in the queue at the intersection; when O = U, ΔT Ui represents the starting delay time of the i-th vehicle in the queue before the upstream intersection, Q U is the average number of vehicles queuing in a single lane on the through lane in the upstream intersection’s coordination direction; when O=D, ΔT Di represents the starting delay time of the i-th vehicle in the queue before the downstream intersection, Q D Coordinate the number of vehicles queued in a single lane on the through lane in the downstream intersection direction; T' is the driver's reaction time; l is the headway between adjacent vehicles, including the length of the preceding vehicle and the safety distance; V' is the velocity of the dissipated wave; B2. The speed change of the vehicle is calculated as follows: Assume that the vehicles after the bth vehicle in the queue are all considered to have the same acceleration as the bth vehicle, and the formula is as follows: Where: O = U, a Ui represents the starting acceleration of the i-th vehicle in the queue at the upstream intersection; when O=D, a Di represents the starting acceleration of the i-th vehicle in the queue before the downstream intersection; f(i) is the acceleration variation law of the vehicle in the queue before vehicle b; B3. According to the kinematic formula, the acceleration time of the queued vehicles is calculated as follows: Where: O=U, T Uai represents the acceleration time of the i-th vehicle in the queue at the upstream intersection; when O=D, T Dai represents the acceleration time of the i-th vehicle in the queue before the downstream intersection; B4. The calculation of the dissipation law of vehicle queues at the downstream intersection is as follows: Where: H k The length of the queue dissipated from the time the green light of the downstream intersection comes on to time k; T is the acceleration of the tail vehicle in the downstream intersection queue; D is the dissipation time of vehicles queued in front of the downstream intersection; Delayed start time for the tail car in the downstream intersection queue; B5. According to the position of the dissipation point, the uniform speed travel time of vehicles at the upstream intersection is calculated as follows: B6. The number of vehicles queued at the downstream intersection shall be based on the data actually detected at different time periods; C. Establishing an adaptive optimization model for green light start time Combined with the vehicle delay at the upstream intersection, the speed change of the vehicle, the number of vehicles in the queue at the downstream intersection and the dissipation law of the vehicles in the queue determined in step B, an adaptive optimization model of the green light start time is established for adjacent intersections and trunk multi-intersections respectively; D. Determine the trunk line adaptive coordination timing optimization scheme based on dynamic green light start time interval In order to coordinate the traffic signals of the W intersections on the trunk line, the cycle durations of the W intersections on the trunk line must be equal; first, according to the layout of each intersection, as well as the traffic flow and direction, the Webster theory is used to calculate the cycle duration required for the traffic signals of each intersection, and then the largest cycle duration is used as the cycle duration of the trunk line adaptive coordinated control, and the intersection with the largest cycle duration is called the key intersection, and the cycle duration of the key intersection is used as the cycle duration of each intersection on the trunk line; combined with the real-time dynamic traffic flow in different time periods, the Webster theory is used to calculate the cycle duration, green-to-signal ratio and green light duration of the W intersections on the trunk line, and finally the dynamic green light starting time distance, i.e., dynamic phase difference, of each adjacent intersection on the trunk line is calculated; Since the trunk adaptive coordinated control involves W intersections, each intersection needs to optimize its own timing optimization scheme according to the dynamic changes of the green light start time interval between adjacent intersections; therefore, the trunk adaptive coordinated timing optimization scheme is evaluated according to the average delay and queue length of each scheme output by the VISSIM software; if the minimum requirements of average delay and queue length are met, go to step E; otherwise, go to step C; E. Output the trunk line adaptive coordination timing optimization plan based on dynamic green light start time interval Output the trunk line adaptive coordination timing optimization plan based on adaptive green light start time distance to provide decision-making information for urban road traffic management personnel.

2. According to claim 1, a trunk line adaptive coordinated control method based on dynamic green light start time interval is characterized in that: The steps of establishing the green light start time adaptive optimization model described in step C are as follows: The green light start time adaptive optimization model includes an adjacent intersection adaptive green light start time model and a trunk line multi-intersection adaptive green light start time model, and the specific steps are as follows: C1. Establishing an adaptive green light start distance model for adjacent intersections According to the influencing factors of the green light start time, after calculating the travel time of each stage of vehicles from upstream to downstream, the adaptive green light start time model of adjacent intersections is established with the goal of allowing vehicles arriving from upstream to pass through the intersection without stopping as follows: Constraints: L D =(Q D -1)·l+l c ,T D =max(ΔT Di +T Dai ) 0≤H p ≤H U,D -L D ΔT Oi ≥0,T Oai ≥0,T Oci ≥0, i=1,2,3,…,Q O ,O=U,D Where: O (U，D) is the green light start time distance at the upstream and downstream intersections; when O=U, ΔT Ui represents the starting delay time of the i-th vehicle in the queue before the upstream intersection, T Uai represents the acceleration time of the i-th vehicle in the queue before the upstream intersection, T Uci represents the uniform speed travel time of the i-th vehicle in the queue before the upstream intersection, Q U is the average number of vehicles queuing in a single lane on the through lane in the upstream intersection’s coordination direction; when O=D, ΔT Di represents the starting delay time of the i-th vehicle in the queue before the downstream intersection, T Dai represents the acceleration time of the i-th vehicle in the queue at the downstream intersection, Q D H is the number of vehicles queuing in a single lane on the through lane in the coordination direction of the downstream intersection; p is the distance between the dissipation point and the last vehicle in the downstream intersection queue; l is the headway between adjacent vehicles; l c is the vehicle length; T D is the dissipation time of the vehicles queuing in front of the downstream intersection; L D H is the queue length of a single lane at the downstream intersection; U，D The distance between the stop lines at the entrances of the upstream and downstream intersections; C2. Establishing an adaptive green light start time model for multiple intersections on trunk roads Assume that there are W intersections on the trunk line, and the intersections are numbered 1, 2, 3, ..., w according to the downstream direction; it is defined that the dissipation is completed when the tail car in the queue at the downstream intersection accelerates to the same speed as the first car arriving at the upstream intersection, that is, the queued vehicles in front of each intersection merge into the coordinated convoy to form a new coordinated convoy. Therefore, when calculating the green light start time between the nth intersection and the n-1th intersection, only the number of vehicles in the queue before the n-1th intersection, the change in vehicle speed and the driving distance need to be considered, and it has nothing to do with the intersections before the n-1th intersection, n = 1, 2, 3, ..., W; then the trunk multi-intersection adaptive green light start time model is regarded as a collection of multiple sets of adjacent intersection adaptive green light start time models, as shown in the following formula: Constraints: L n =(Q n -1)·l+l c ,T n =max(ΔT nj +T naj ) 0≤H np ≤H n-1,n -L n ΔT (n-1)i ≥0,T (n-1)ai ≥0,T (n-1)ci ≥0, i=1,2,3,...,Q n-1 ,j=1,2,3,...,Q n Where: O (n-1，n) is the green light start time interval between the n-1th intersection and the nth intersection, where n = 1, 2, 3…w; ΔT (n-1)i is the starting delay time of the i-th vehicle in the queue before the n-1-th intersection; T (n-1)ai V is the acceleration time of the i-th vehicle in the queue before the n-1-th intersection; nf is the uniform speed of vehicles on the road between the n-1th intersection and the nth intersection; a (n-1)i is the acceleration of the i-th vehicle in the queue before the n-1-th intersection; T (n-1)ci is the uniform speed travel time of the i-th vehicle in the queue before the n-1-th intersection; H n-1，n is the distance between the n-1th intersection and the nth intersection; H np is the distance between the dissipation point before the nth intersection and the last vehicle in the queue; T n is the dissipation time of the queued vehicles before the nth intersection; Q n is the number of vehicles queued before the nth intersection; L n is the queue length before the nth intersection.

3. According to claim 1, a trunk line adaptive coordinated control method based on dynamic green light start time interval is characterized in that: The method for determining the trunk line adaptive coordination timing optimization scheme based on the dynamic green light start time distance described in step D is as follows: In the process of arterial adaptive coordinated control, for the g cycle, the cycle duration of this cycle is determined by the cycle duration of the key intersection in the g-1 cycle; obviously, when using the traffic flow data obtained in the g-1 cycle to calculate the cycle duration of the g cycle, it is necessary to ensure that before the traffic flow of the g-1 cycle is obtained, the signal timing of the starting intersection in the arterial coordinated control is still executing the timing scheme of the g-1 cycle, that is, the signal timing of the starting intersection has not entered the g cycle; After obtaining the road traffic flow data of the key intersection in the g-1 cycle of the trunk adaptive coordinated control, the optimal cycle duration required for the key intersection is calculated as the cycle duration of the g cycle: Where: C g is the cycle length of the g cycle; L is the total loss time of the key intersection of the g-1 cycle; Y is the sum of the maximum traffic flow ratios of each phase of the key intersection of the g-1 cycle; The total loss time and the sum of the maximum traffic flow ratios of each phase at the intersection are: Where: l' is the start-up loss time; A is the yellow light time; I is the green light interval time, that is, the time from the end of the green light of the previous phase to the beginning of the green light of the next phase; u is the number of phases in the g-1 cycle; Where: y u is the traffic flow ratio of the critical lane in the uth phase of the critical intersection of the g-1 cycle; In the coordinated control process of the arterial line, the allocation of the green light time of each phase of the intersection is also based on the principle of minimizing vehicle delays. Therefore, the effective green light time of each phase should be allocated according to the flow ratio of each phase. For any nth intersection, according to the measured traffic flow Q of each entrance road in different directions in the g-1 period, xy After that, the traffic flow ratio y of the critical lane in phase u within the g-1 period is calculated nu ; The effective green light time G of the uth phase in the gth cycle of the nth intersection eun The calculation formula is as follows: Where: G enu is the effective green light time of the uth phase in the gth cycle of the nth intersection; y nu is the traffic flow ratio of the u-th phase critical lane in the n-th intersection within the g-1 cycle; C g is the cycle length of the g cycle; L n is the total loss time of the nth intersection; Y n is the sum of the maximum traffic flow ratios of each phase at the nth intersection in the g-1 period; The green-to-signal ratio of each phase in the g-cycle of the n-th intersection is: Where: nu is the green-to-signal ratio of the u-th phase in the g-th period of the n-th intersection; The green light time for each phase is: G nu =G enu -A u +l u ' Where: G nu A is the green light display time of the uth phase in the gth cycle of the nth intersection; u is the yellow light time of the uth phase; l′ u It is the startup loss time of the uth phase.

Images