CN110889967B - An overflow risk balance signal control optimization method based on trunk road segmentation - Google Patents

Abstract

The invention discloses a method for overflow risk balance signal control optimization based on trunk road segmentation. The method includes: using two indicators, the periodic queuing clearing area and the overflow risk area, to identify the overflow status of the intersection; and then using the arterial road segmentation method to divide the arterial road into a collection of overflow risk prevention and control sub-areas, each sub-area contains three types of Different types of intersections: input, output and connecting intersections; next, for different types of intersections in the sub-area, a signal control optimization strategy of overflow risk balance is adopted for the straight phase of the arterial road, and then the left direction of the arterial road is adjusted according to the queue length of each phase. The rotation phase and branch phase are used for signal control optimization to determine the green light duration of each phase.

Description

An overflow risk balance signal control optimization method based on trunk road segmentation

Technical field

This application relates to the field of traffic signal control, specifically, to an overflow risk balance signal control optimization method based on trunk road segmentation.

Background technique

With the explosive growth of urban traffic demand, urban arterial roads have become increasingly congested, leading to frequent queue overflows at intersections. Nowadays, "Internet + signal light" has become a hot research topic. Alibaba released "City Brain 2.0" to take over the timing optimization of 1,300 traffic lights in Hangzhou; Didi Chuxing used crowdsourced trajectory data such as taxis and express trains to build a "perception-evaluation-optimization-implementation" closed-loop traffic signal monitoring and optimization The system has implemented more than 1,500 sets of intersection timing optimization solutions in cities and regions such as Liuzhou, Jinan, and Beijing Capital Airport, which has alleviated traffic congestion problems to a certain extent.

However, the above practices only achieve "coarse-grained" traffic status assessment regardless of lanes and flow directions and "multi-period" periodic signal timing optimization for Time Of Day (TOD) division. In addition, most optimization controls passively respond to traffic demand changes, and lack in-depth analysis of the evolution process and control strategies of queuing from normal state to overflow state. Considering that the traffic demand on urban arterial roads continues to be large, intersections are often at high risk and high frequency of queuing overflow.

Queue length is a more intuitive indicator that depicts the traffic operating status at the spatial level, and the direct manifestation of queue overflow is the generation of excessively long queues. Therefore, how to evaluate the traffic status in a "fine-grained" manner based on the queue length, period and flow direction, and timely , Accurately predict the risk of queuing overflow, and establish an optimization theory and method system for active prevention of queuing overflow and rapid risk elimination, which are key theoretical issues that need to be solved urgently in the field of traffic control.

Contents of the invention

1. Purpose of invention

In order to solve the problem that the current signal control method cannot identify the queuing overflow risk in a timely and accurate manner, and the signal control strategy is relatively lagging in responding to traffic demand, this invention proposes an overflow risk balance signal control optimization method based on trunk road segmentation to eliminate trunk road overflow. flow risk.

2. Technical solutions adopted in the present invention

The overflow risk balance signal control optimization method based on trunk road segmentation proposed by this invention can be realized through the following steps:

(1) Based on the straight phase queue length of each intersection on the main road, calculate two indicators: the intersection periodic clearing area and the overflow risk area;

(2) Based on the periodic clearing area and overflow risk area calculated in step (1), identify the overflow risk status of the intersection and divide the overflow risk status of the intersection into three categories: low, medium and high overflow risk ;

(3) According to the identification results in step (2), the adjacent intersections with medium or high overflow risk are included in the same overflow sub-area, so as to divide the main road into different overflow sub-areas. At the same time, each An overflow sub-area usually contains three types of intersections: input, output and connecting intersections;

(4) For different types of intersections within the overflow sub-area, signal control optimization strategies such as flow limiting, balancing, and maximum flow are adopted to achieve a balance of straight phase overflow risks at each intersection on the main road.

The step (1) is specifically as follows: Since each intersection section of the main road accommodates a large number of vehicles, the poor signal control effect of one cycle is not enough to cause the queuing overflow phenomenon, and the queuing overflow phenomenon is usually caused by Accumulated results over 2 cycles. Therefore, how to accurately identify the intersection overflow risk during the evolution of the intersection from a saturated state to an overflow state is the key to the design of the overflow prevention signal control scheme. This paper proposes two periods, the periodic queuing clearing area q _cr and the overflow risk area s _cr . An indicator is used to identify the overflow risk of the intersection in each cycle, and the trunk road is segmented according to the overflow risk of each intersection.

The periodic queue clearing area is an indicator to evaluate whether the intersection is in a saturated state. The length is first related to the traffic capacity of the intersection within the period. If the queue length is greater than the cycle capacity, the queued vehicles within the cycle cannot be cleared, and an initial queue will be generated at the beginning of the next cycle, leaving the intersection in a saturated state. Queue clearing capacity can be calculated by the following formula:

q _max =q _s g _k (1)

Among them, q _s is the saturated flow rate of the intersection (vehicles/second). When the green light starts, the queuing vehicles in the cycle pass through the intersection at the saturated flow rate, and g _k is the green light duration of the kth cycle of the intersection.

The length of the periodic queue clearing area is secondly related to the coordination between this intersection and the upstream intersection. After the green light starts at the upstream intersection, the queuing vehicles at the upstream intersection enter this intersection in a convoy. If the queuing vehicles at this intersection have been cleared at this time, the upstream convoy will directly pass through the intersection before the green light ends; if the queuing vehicles at this intersection have cleared If the queuing vehicles have not completely dissipated when the upstream fleet arrives, the upstream fleet will join the queue. Due to the large traffic flow on the main road, the queue length will increase rapidly, creating the risk of overflow. The queue length threshold considering upstream intersection coordination can be calculated by the following formula:

where v ₂ is the evanescent wave velocity (m/s), v _f is the free flow velocity (m/s), k _j is the congestion density (number of vehicles/m), is the green light start time of the kth intersection, D is the distance from the parking line of the upstream intersection to the parking line of this intersection, T _coor and q _coor are the generation time and queue length threshold of the queue length threshold respectively.

Combining the queue clearing capacity and queue length threshold of this intersection, the periodic queue clearing area of this intersection can be determined using the following formula:

q _cr =min(q _max ,q _coor ) (4)

The overflow risk area is located at the end of the road section and is an indicator to determine whether the intersection is in a high overflow risk state. The range of the overflow risk area is proportional to the length of the road section. The longer the road section, the larger the overflow risk area, and conversely, the smaller the overflow risk area. The calculation formula for the overflow risk area can be obtained from the following formula:

s _cr = β _cr Lk _j (5)

Among them, β _cr is the overflow area coefficient, k _j is the congestion density (vehicles/meter), and L is the length of this road section (meters).

The specific step (2) is: segment the main road in each cycle. First, it is necessary to identify the overflow risk status of the straight phase of each intersection on the main road in each cycle. Based on the queuing clearing area and overflow risk area introduced above, it can be The overflow risk status of the intersection is divided into three categories: low overflow risk, medium overflow risk and high overflow risk. The classification principles are as follows:

Among them, q _total = k _j L is the total number of vehicles that the road section can accommodate, and L is the length of this road section.

For signalized intersections with low overflow risk during the cycle, it means that the signal control effect is better during this cycle and traffic supply and demand are balanced. There is no need to optimize the signal control in the next cycle, and the traffic on the adjacent intersection on the arterial road can be appropriately reduced. Demand-oriented intersections are at low overflow risk; for signalized intersections at medium overflow risk during the cycle, it means that the intersection is in a saturated state, traffic demand is greater than traffic supply, and intersection signal control needs to be appropriately optimized to prevent the continuous flow of queuing vehicles. The accumulation causes queuing overflow phenomenon; for signalized intersections with high overflow risk during the cycle, it means that the intersection demand is far greater than the intersection supply, and the intersection has already produced or is about to produce queuing overflow phenomenon, and signal control optimization needs to be carried out in a timely manner. By limiting traffic inflow and increasing traffic supply, we can achieve a balance between supply and demand at the intersection and prevent queue overflow in the next cycle.

The specific step (3) is: according to the overflow risk identification results in step (2), adjacent intersections with medium and high overflow risks are included in the same overflow sub-area, so that the main road is divided into Multiple overflow sub-zones. Each overflow subzone contains three types of intersections: input intersections, output intersections, and connecting intersections. The intersection upstream of the most upstream overflow intersection in the overflow sub-area is called the input intersection. The traffic flow at the upstream intersection of the overflow sub-area flows into the interior of the overflow sub-area through the input intersection; the output intersection is the most upstream intersection in the overflow sub-area. For downstream intersections, the traffic flow within the overflow sub-area is output to the downstream section of the overflow sub-area through the output intersection; connecting intersections are all intersections between the input intersection and the output intersection, and the traffic flow within the overflow sub-area is of traffic flows downstream through connecting intersections.

The step (4) is specifically:

The straight phase of urban arterial roads is the main direction of traffic flow on arterial roads. Therefore, the key to anti-overflow signal control on arterial roads is to prevent queuing overflow in the straight phase of arterial roads. Signal control optimization of the straight phase adopts different signal control optimization methods according to different intersection types in the overflow sub-area.

The input intersection is the most upstream intersection within the overflow sub-area, and external traffic flow flows into the overflow sub-area through the input intersection. Since the straight phase of the input intersection itself is in a non-overflow state, and the intersections within the overflow sub-area are in a saturated state, the goal of input intersection signal control is to limit the flow of traffic into the overflow sub-area by reducing the green light duration of the straight phase. Prevent queuing overflow from flowing into the overflow sub-area due to too many vehicles. The reduction in the green light time of the straight phase at the input intersection reduces the traffic supply, and the overflow risk is directed from the inside of the overflow sub-area to the upstream intersection, which plays a role in balancing the overflow risk at each intersection on the main road.

The reduction of the green light duration at the input intersection is determined by the queue length of the downstream straight phase. The calculation formula for the green light duration in the next cycle of the straight phase at the input intersection is as follows:

in is the green light duration of the k-th cycle at the l-th intersection, l+1 is the downstream intersection of the l-th intersection, The length of the queue clearing area for the kth cycle of the lth intersection, the min(A,B) function returns the smaller value of A and B,/> is the remaining vehicle capacity in the straight phase section of the l-th intersection in the k-th cycle, and K _i1 and K _i2 are the incremental coefficients of the input intersection.

The output intersection is the most downstream intersection in the overflow sub-area, and the traffic flow within the overflow sub-area flows out of the overflow sub-area through the output intersection. The downstream intersection of the output intersection is in a non-overflow state, so the goal of signal control of the output intersection is to maximize the traffic supply of the output intersection by increasing the green light duration of the straight phase, so that the overflow risk is directed to the downstream intersection, thereby Balance the risk of overflow at each intersection on the main road. The calculation formula for the increase in the green light duration at the output intersection is similar to formula (7). The increase in the green light duration is determined by the excess of the queue length beyond the queue clearing area and the change in the queue length in the overflow risk area. The formula is as follows:

Among them, K _o1 and K _o2 are the incremental coefficients of output intersection signal control, and the definitions of the other parameters are shown in formula (7).

The connecting intersection is located between the input intersection and the output intersection of the overflow sub-area, and the traffic flow within the overflow sub-area propagates downstream through the connecting intersection. Since all connecting intersections are in a medium-to-high overflow risk state, the goal of signal control at connecting intersections is to adjust the green light duration of the straight phase of each connecting intersection so that the overflow risk is at a lower level within the overflow sub-area. Intersection propagation of risk status balances the overflow risk of each intersection within the overflow sub-area.

The change in green light duration of the straight-going phase of each connecting intersection in the next cycle is determined by the overflow risk of this intersection and the downstream intersection. If the overflow risk of this intersection is higher than that of the downstream intersection, then the straight-going phase of this intersection in the next cycle The green light duration increases; if the overflow risk level of this intersection is lower than that of the downstream intersection, then the green light duration of the straight phase of this intersection will decrease in the next cycle. The calculation formula for the periodic straight phase green light duration at the connecting intersection is as follows:

Among them, K _c1 and K _c2 are the incremental coefficients connecting the intersection signal control, and the definition of the remaining parameters is shown in formula (7).

After determining the green light duration of the straight phase of the arterial road in the next cycle, it is necessary to determine the green light duration of the left turn phase of the arterial road in the next cycle. This algorithm first determines the green light duration for the left turn in the next cycle through the queue length of the left turn phase within the cycle on the premise of ensuring the green light duration for the straight phase of the arterial road. The calculation formula is as follows:

Among them, A and B are the relative traffic flow directions of the main road (eastward and westward in this article), T and L represent the straight phase and the left turn phase respectively, g and q represent the green light duration and the queue length. therefore Represents the duration of the left-turn green light in direction A of the k-th period of the i-th intersection,/> Represents the queue length of the left turn phase in direction A of the kth cycle at the i-th intersection. /> is the length of the partition on one side of the main road in the double-ring structure (that is, the total green light time for two-phase traffic on the same ring on the main road), G _max is the maximum green light time allowed for two-phase traffic on the main road, q _s is the saturated flow rate of the intersection (vehicles/second).

The main purpose of the signal control optimization method in this article is to prevent the overflow phenomenon on the main road. The traffic demand of the branch road is much lower than the traffic demand of the main road. However, it is also necessary to ensure the basic traffic capacity of the branch road and avoid the overflow phenomenon of the branch road due to insufficient traffic supply. When the total green light duration of the trunk road is determined, the branch signal control method allocates the remaining green light duration in the next cycle based on the queue length ratio of the branch phase. The calculation formula for the branch phase green light duration is as follows:

where C is a fixed period length. Since it is assumed that there is no signal overlap phenomenon in the branch, cs1 and cs2 respectively represent different branch phases.

To sum up, the pseudo code of the overflow risk balance signal control optimization algorithm based on trunk road segmentation proposed by this invention is as follows:

Description of the drawings

The drawings described here are used to provide a further understanding of the present application and constitute a part of the present application. The illustrative embodiments of the present application and their descriptions are used to explain the present application and do not constitute an improper limitation of the present application. In the attached picture:

Figure 1 is a satellite map of an actual arterial road network according to specific embodiments of the present disclosure;

Figure 2 shows the VISSIM simulated road network and initial signal timing according to the specific implementation of the present disclosure;

Figure 3 is an example diagram of an arterial road segmentation method according to specific embodiments of the present disclosure;

Figure 4 is a vehicle trajectory diagram of the eastbound straight-bound phase of the main road before signal optimization according to the specific embodiment of the present disclosure;

Figure 5 is a vehicle trajectory diagram of the eastbound straight-bound phase of the arterial road after signal optimization according to the specific embodiment of the present disclosure.

Figure 6 is a main flow diagram of a method according to an embodiment of the present disclosure.

Detailed ways

The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are only some of the embodiments of the present invention, rather than all the embodiments. The following description of at least one exemplary embodiment is merely illustrative in nature and is in no way intended to limit the invention, its application or uses. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without making creative efforts fall within the scope of protection of the present invention.

It should be noted that the terms used herein are only for describing specific embodiments and are not intended to limit the exemplary embodiments according to the present application. As used herein, the singular forms are also intended to include the plural forms unless the context clearly indicates otherwise. Furthermore, it will be understood that when the terms "comprises" and/or "includes" are used in this specification, they indicate There are features, steps, operations, means, components and/or combinations thereof.

The relative arrangement of components and steps, numerical expressions, and numerical values set forth in these examples do not limit the scope of the invention unless specifically stated otherwise. At the same time, it should be understood that, for convenience of description, the dimensions of various parts shown in the drawings are not drawn according to actual proportional relationships. Techniques, methods and equipment known to those of ordinary skill in the relevant art may not be discussed in detail, but where appropriate, the techniques, methods and equipment should be considered part of the authorized specification. In all examples shown and discussed herein, any specific values are to be construed as illustrative only and not as limiting. Accordingly, other examples of the exemplary embodiments may have different values. It should be noted that similar reference numerals and letters refer to similar items in the following figures, so that once an item is defined in one figure, it does not need further discussion in subsequent figures.

The overflow risk balance signal control optimization method based on trunk road segmentation proposed by this invention can be realized through the following steps:

(1) Based on the straight phase queue length of each intersection on the main road, calculate two indicators: the intersection periodic clearing area and the overflow risk area;

(2) Based on the periodic clearing area and overflow risk area calculated in step (1), identify the overflow risk status of the intersection and divide the overflow risk status of the intersection into three categories: low, medium and high overflow risk ;

(3) Based on the identification results in step (2), adjacent intersections with medium or high overflow risk status are included in the same overflow sub-area, thereby dividing the main road into different overflow sub-areas. At the same time, each overflow subzone usually contains three types of intersections: input, output, and connecting intersections;

(4) For different types of intersections within the overflow sub-area, signal control optimization strategies such as flow limiting, balancing, and maximum flow are adopted to achieve a balance of straight phase overflow risks at each intersection on the main road.

The specific methods of each step are introduced below.

Step (1): In order to simulate the actual trunk road network, this case uses the microscopic traffic software VISSIM to build a simulation road network model. As shown in Figure 1, this paper selects six consecutive signalized intersections on Jingshi Road in Jinan City as the research object. Jingshi Road is an important east-west traffic artery in Jinan City. The traffic flow, especially the morning and evening peak traffic flow, is large and the trunk road is long. The time is in a saturated state and it is easy to cause queue overflow. The six intersections selected in this part are Shungeng Road, Qianfoshan Road, Lishan Road, Shanshi East Road, Huanshan Road and the intersection of Shanda Road and Jingshi Road from west to east. The total length of each section of the intersection is 2.8 km, the road speed limit is 50km/h, the initial signal timing plan of each intersection is shown in Figure 2, and the cycle length is fixed at 210s.

This case assumes that the queue length of each intersection on the main road can be accurately obtained as input. First, formulas (1)-(4) are used to calculate the two indicators of the periodic queue clearing area of each intersection and the overflow risk area of formula (5), where overflow The risk area coefficient is set to 0.2;

Step (2): As shown in Figure 3, taking six consecutive west-to-east intersections of Jingshi Road in Jinan City as an example to further illustrate how to implement the arterial road segmentation algorithm within a cycle. During this period, formula (6) was used to identify the overflow risk status of each intersection on the main road. It was found that the Huanshan Road intersection was in a low overflow risk status, the Qianfoshan Road intersection was in a medium overflow risk status, and the Lishan Road and Shanshan Road intersections were in a medium overflow risk status. The intersection of Shidong Road and Shanda Road is at high risk of overflow;

Step (3): Based on the identification results in step (2), after incorporating adjacent intersections with medium and high overflow risks into the same overflow sub-area, the periodic trunk road is divided into two overflow sub-areas. area, as shown in Figure 2. Among them, overflow sub-area 1 uses Shungeng Road, the upstream intersection of Qianfoshan Road, the first overflow intersection, as the input intersection, and Shanshi East Road, the most downstream, as the output intersection. Two intersections (Qianfoshan Road and Lishan Road) serve as connecting intersections. Overflow sub-area 2 uses Huanshan Road and Shanda Road as input and output intersections respectively. Since there are no other intersections between the input and output intersections, overflow sub-area 2 has no connecting intersections;

Step (4): For different intersections within the overflow sub-area, formulas (7), (8), and (9) are used to optimize the straight-going phase signal control, where the incremental coefficient values are K _i1 = 0.15. , K _i2 =0.8, K _o1 =0.25, K _o2 =0.8, K _c1 =0.1, K _c2 =0.5. After determining the green light duration of the straight phase of the arterial road, use formulas (10) and (11) to determine the green light duration of the left turn phase and branch phase of the arterial road respectively.

Figure 3 shows the spatio-temporal trajectories of all vehicles in one lane in the straight phase at each intersection in the east direction of the main road before and after 20 cycles of signal optimization. As can be seen from Figure 4, the main road was in a saturated state before signal optimization. Due to the imbalance of traffic supply and demand at the intersections, the overflow phenomenon at the three intersections of Qianfoshan Road, Lishan Road and Shanshi East Road was serious, and there would be an overflow phenomenon every cycle. Queue overflow phenomenon seriously affects the operational efficiency of intersections. As shown in Figure 5, after adopting the signal control optimization algorithm proposed in this chapter, the queuing and overflow phenomenon at the intersection is significantly reduced. It can be seen that the overflow risk balance signal optimization algorithm based on trunk road segmentation can effectively reduce the overflow risk of trunk roads.

In this embodiment, the overflow risk balance signal control optimization method based on trunk road segmentation is used. The simulation results show that after the saturated trunk road is optimized by the signal control optimization algorithm proposed in this chapter, the incidence of queuing and overflow phenomena on the trunk road is significantly reduced, and the trunk road traffic is improved. Efficiency is significantly improved.

Claims (2)

1. A method for optimal control of overflow risk balance signals based on trunk road segmentation, which is characterized by including: The two indicators of periodic queuing clearing area and overflow risk area are used to identify the overflow status of the intersection; the arterial road is divided into a collection of overflow risk prevention and control sub-areas through an arterial road segmentation method based on overflow risk clustering. Use the periodic clearing area and overflow risk area indicators to identify the overflow risk status of the intersection, including low, medium and high overflow risk, and include adjacent intersections with medium or high overflow risk status into the same overflow Sub-area, the main road is divided into different overflow sub-areas. Each sub-area contains three different types of intersections: input, output and connecting intersections. The upstream intersection of the most upstream overflow intersection in the overflow sub-area is called It is an input intersection, and the traffic flow at the upstream intersection of the overflow sub-area flows into the overflow sub-area through the input intersection; the output intersection is the most downstream intersection of the overflow sub-area, and the traffic flow inside the overflow sub-area passes through the output intersection. The intersection outputs to the downstream section of the overflow sub-area; connecting intersections are all intersections between the input intersection and the output intersection, and the traffic flow within the overflow sub-area propagates downstream through the connecting intersections; For different types of intersections in the sub-area, the overflow risk balance signal control optimization strategy is adopted for the straight phase of the arterial road. Specifically, it reduces the input of the input intersection, increases the output of the output intersection, and adjusts the flow between each connecting intersection, so that The overflow risk propagates within the overflow sub-area to intersections with lower overflow risk. Then, on the premise of ensuring the green light duration of the straight phase of the main road, the left turn phase of the next cycle is determined by the queue length of the left turn phase within the cycle. The green light duration is determined, and when the total green light duration of the trunk road is determined, the remaining green light duration in the next cycle is allocated based on the queue length ratio of the branch phase. 2. The method according to claim 1, characterized in that, based on the straight phase queue length of each intersection on the main road, the operation of calculating the two indicators of the intersection periodic clearing area and the overflow risk area includes: The periodic queue clearing area is calculated based on the queue clearing capacity of the intersection and the queue length threshold considering the coordination of the upstream intersection, and the overflow risk area is calculated based on the length of the road section.