CN111145565B - Method and system for recommending coordination route and coordination scheme for urban traffic - Google Patents

Abstract

The invention relates to a method and a system for recommending a coordinated route and a coordinated scheme for urban traffic. The output result of the invention can be used for the evaluation of the coordinated route recommendation, and can also be used for the output of the coordinated route recommendation scheme and the implementation of the actual urban traffic road section. By carrying out real-time coordination control on the urban traffic trunk line, the urban traffic flow can smoothly flow, and the modernization construction and the improvement of the urban comprehensive traffic control capacity are promoted.

Description

Method and system for recommending coordination route and coordination scheme for urban traffic

Technical Field

The invention relates to the field of urban traffic, in particular to a method and a system for recommending a coordination route and a coordination scheme for urban traffic.

Background

In an urban road network, a large number of intersections are close to each other. The inconsistent signal timing of adjacent intersections often results in vehicles encountering green lights when passing through upstream intersections and red lights when arriving at downstream intersections, thereby causing a large-scale queuing phenomenon. Due to the short distance between intersections, the distance between two adjacent intersections is usually not sufficient to completely evacuate the traffic stream, and the vehicles in line extend from the downstream intersection to the upstream, leaving the entire traffic system in a paralyzed state. Therefore, the vehicles stop and start at the intersection, so that the driving is not smooth, the road network operation efficiency is low, and the problems of aggravation of environmental pollution, energy waste and the like are caused.

In order to reduce the number of times of stopping vehicles at the intersection and the driving delay, a traffic control strategy needs to be implemented at the intersection on the trunk line. The control strategy adopted in China at present has the following defects: the method is classified according to the control principle, and each parameter needs to be manually set by a timing control strategy, so that the efficiency is not high; the induction control needs to arrange a vehicle detector at an entrance road of an intersection, and the control mode is changed at any time according to traffic information detected by the detector, but the induction control is influenced by the actual detector equipment condition and road condition, and often has a lot of data missing conditions. According to the control range, most of the intersections in China currently adopt single-point control, the traffic control signal of each intersection only operates independently according to the traffic condition of the intersection and is not in any connection with the control signals of other adjacent intersections, but the control strategy can only be applied to unsaturated traffic flow due to the fact that the actual adjacent intersections are close in distance, and efficiency is not high. In addition, a coordination control mode is adopted for part of main traffic lines, the method can be suitable for adjacent intersections with close distances, a fleet on the main lines forms green wave bands, and one line passes through the main lines without resistance, but the coordination control sacrifices the passing capacity of branch lines and is not beneficial to the traffic condition of the whole road network. Therefore, in the face of excessively complex factors or emergencies, the traditional traffic signal control method cannot meet the requirements of current traffic control optimization. And under the condition that the traffic flow rule changes greatly, parameters of signal timing of the signal control system generally need to be adjusted manually in real time. The regulation and control mode has the defects of non-reproducibility, low efficiency, low reliability and the like, and a novel technology is urgently needed to be used as an auxiliary means to alleviate the problems.

Disclosure of Invention

The invention aims to overcome the defects and provides a method and a system for recommending a coordinated route and a coordinated scheme for urban traffic. The output result of the invention can be used for the evaluation of the coordinated route recommendation, and can also be used for the output of the coordinated route recommendation scheme and the implementation of the actual urban traffic road section. By carrying out real-time coordination control on the urban traffic trunk line, the urban traffic flow can smoothly flow, and the modernization construction and the improvement of the urban comprehensive traffic control capacity are promoted.

The invention achieves the aim through the following technical scheme: a method for recommending a coordination route and a coordination scheme for urban traffic comprises the following steps:

(1) collecting multi-source heterogeneous traffic data of a road network, and preprocessing the data;

(2) acquiring a coordination route, selecting a target road section from the coordination route, and establishing a state equation of the flow relation of the target road section changing along with time;

(3) extracting state parameters, control parameters, state time-varying parameters and control time-varying parameters in a state equation of the target road section, constructing an optimal control system of the target road section based on a DDALQR algorithm, and solving optimal time-varying parameters of the target road section, wherein the optimal time-varying parameters describe optimal current-time control parameters under current-time state parameters;

(4) calculating the coordinated green light time of the coordinated intersections at the two ends of the target road section at the current moment according to the optimal time-varying parameter of the target road section and the state of the target road section at the current moment;

(5) respectively calculating a coordination period, a coordination phase difference, a coordination cycle and a coordination scheme of the coordination route based on the coordination route;

(6) the coordination route and the effect of the coordination scheme thereof are evaluated.

Preferably, when the coordination route is acquired, the coordination route is recommended based on the determined trunk road, and the intersection of the trunk road is divided into a plurality of groups of coordination routes according to a threshold judgment method; the coordination route comprises two or more continuous coordination intersections, a road section between two adjacent continuous coordination intersections and a coordination direction between two adjacent continuous intersections; the main road intersection is divided into a plurality of groups of coordination routes according to a threshold judgment method, and the following steps are specifically performed:

(i) acquiring initial flow ratio data of a week from a database, setting a flow ratio threshold condition and a distance threshold condition, and acquiring upstream and downstream relation data of a road section and the maximum value and the minimum value of coordinated intersection data;

(ii) extracting all intersections or road sections meeting the threshold condition, and then judging all data meeting the condition;

(iii) judging whether the downstream crossing of the crossing meeting the conditions meets the threshold condition and the maximum and minimum limit conditions of the coordinated crossing, adding the downstream crossing meeting the conditions to the crossing coordinated road route crossing list, and discarding the coordinated crossing list if the downstream crossing does not meet the conditions; screening the coordinated intersections, segmenting the main road intersections which cannot be coordinated, and storing results to a database;

(iv) acquiring the calculated coordination route information;

(v) setting new flow ratio data or a coordination route under the condition of threshold value condition, and calling a function to return new data.

Preferably, the establishing of the state equation in the step (2) is specifically as follows: let us choose the target link L from the coordinated route numbered j_i，jEstablishing a target road section L_i，jThe equation of state of (c):

wherein the target section L is flowed into_i，jThe flow rates of (a) come from: road section L_i，j+1Left-turn lane, road section L_i，j+2Straight lane, road section L_i，j+3Right-turn lane of (1), target road section L at next moment_i，jFlow rate x of_i，j(k +1) is the current flow x_i，j(k) Adding the flow flowing into the road section at the current moment, and subtracting the flow flowing out of the road section at the current moment; the equation of state is shown below:

x_i，j(k+1)＝x_i，j(k)+S_i，j+1(k)Δg_i，j+1(k)+S_i，j+2(k)Δg_i，j+2(k)+S_i，j+3(k)Δg_i，j+3(k)-S_i，j(k)Δg_i，j(k)

k represents the kth time; s_i，j+1(k)，S_i，j+2(k)，S_i，j+3(k)，S_i，j(k) Respectively a road section L_i，j+1Left-turn lane, road section L_i，j+2Straight lane, road section L_i，j+3The saturation flow rate of the right-turn lane and the target road section at the moment k is the ratio of the maximum value of the historical flow rate to the green light time; Δ g_i，j+1(k)，Δg_i，j+2(k)，Δg_i，j+3(k)，Δg_i，j(k) Respectively a road section L_i，j+1Left-turn lane, road section L_i，j+2Straight lane, road section L_i，j+3The right-turn lane and the green light change time of the target road section at the time k.

Preferably, the step (3) specifically includes:

(3.1) extracting state parameters, control parameters, state time-varying parameters and control time-varying parameters in the state equation of the target road section;

(3.2) constructing an optimal control system of the target road section based on a DDALQR algorithm;

and (3.3) obtaining the optimal time-varying parameter of the target road section, wherein the optimal time-varying parameter describes the optimal current-time control parameter under the current-time state parameter.

Preferably, the step (4) is specifically:

if the intersection I_iIntersection I_i+2Intersection I_i+3Coordination exists, and the calculation process of the coordination green light time is as follows:

(a) according to the coordinated direction, the junction I is calculated by using the Webster method_iThe optimal green light time scheme is used as an initialization green light time scheme;

(b) calculation of coordinated intersection I by using DDALQR algorithm_i+1A coordinated green time scheme of (1);

(c) calculating the knot according to step (b)Fruit, update intersection I_i+1The green light time scheme distributes the updated scheme in proportion as the intersection I_i+1The optimal green light time coordination scheme is adopted;

(d) based on DDALQR algorithm, updated intersection I is utilized_i+1Intersection I coordinated with green light time scheme calculation_i+2The optimal coordination green light time scheme is as follows: along the coordination direction of the coordination route, each coordination intersection carries out the operation until the target road section of the last coordination intersection is operated; and acquiring the flow x (k) of the target road section at the current moment, acquiring an optimal control vector according to the optimal time-varying parameter of the target road section, and finishing the split horizon control of the DDALQR algorithm.

Preferably, the step (5) is specifically:

(5.1) calculating the coordination period: calculating each coordination period of the coordination route according to saturation data of the coordination intersection based on the determined coordination route;

(5.2) calculating a coordination phase: calculating all phases which are possible to coordinate according to the corresponding relation between the corresponding road section of the coordinated intersection and the lamp group and the corresponding relation between the lamp group and the phase, namely the coordinate phase; in the same time period, a single intersection can only appear in one group of coordination routes, and the coordination routes in the same time period can only reserve one group of coordination routes with the largest number of intersections;

(5.3) calculating the coordination route phase difference: for the calculation of the signal period of the intersection I, the calculation formula is as follows:

wherein l_IDistance between the upstream intersection and the downstream intersection; v. of_IThe average speed of the vehicle between the upstream intersection and the downstream intersection;

in order to control the parameters of the device,the value is continuously calculated to obtain an optimal result; v. of^cIs constant and is determined according to the actual traffic equipment data; x is the number of_L(k) The traffic flow in the 1h period on the road section L;

the maximum traffic flow in the historical period on the road section L;

(5.4) calculating a coordination period: adjusting the intersection signal period through a feedback algorithm; the algorithm adjusts the signal cycle time of the intersection to enable the intersection cycle to adapt to the maximum saturation of the intersection in the road network, wherein the intersection cycle updating method comprises the following steps:

for intersection I, the calculation formula of the intersection signal period is as follows:

C(k)＝C^N+K^C(σ(k)-σ^N)

σ_I(k)＝x_L/x_L ^max

wherein C (k) is the intersection signal period at the moment k; c^NThe period duration of the existing signal; k^CThe variable control parameters have values which influence the intensity of traffic control, and the values are averaged in the routes which meet the coordination requirement according to multiple calculations; sigma_I(k) The current load of the intersection; x is the number of_LThe traffic flow in the period of the road section L;

the maximum traffic flow in the road section period is obtained; sigma^NThe average load of the coordinated route is the average value of the loads of all intersections under the coordinated route;

(5.5) calculating a coordination scheme: selecting a target road section according to the coordination direction of the coordination route, and calculating a coordination scheme of the coordination route; and (4) distributing the coordinated green light time obtained by the calculation in the step (4) to respective phases according to the original phase distribution proportion, thereby completing the green signal ratio control and the period control of the coordinated route.

Preferably, the coordination route and the effect of the coordination scheme thereof are evaluated, wherein the evaluation comprises a delay index and a travel shortening time, and the delay index of the coordination route is calculated as follows:

the delay index of the road is the ratio of the real-time passing time of the passing route to the passing time of the passing route in a smooth state, and is calculated according to the following formula:

d＝t₁/t₂

wherein d is a delay index of the road; t is t₁The real-time road passing time of the route is the unit of second; t is t₂The time of passing through a route in a smooth state is expressed in seconds; the real-time passing time of the passing route refers to the ratio of the distance of the passing route to the design speed of the route under the condition of free flow, namely, the real-time passing time is calculated according to the following formula:

t₁＝∑s/∑v₁

wherein, t₁The real-time road passing time of the route is the unit of second; s is the distance through the route in km; v. of₁The unit is km/h of the designed speed of a passing route; under the condition of free flow, the design speeds of all road sections of the route are different, and then summation is carried out during subsection calculation of different sections on the route;

the real-time road passing time of the passing route refers to the ratio of the distance of the passing route to the design speed of the route in a smooth state, namely the real-time road passing time is calculated according to the following formula:

t₂＝∑s/∑v₂

wherein, t₂The time is the time of passing through a route in a smooth state, and the unit is second; s is the distance through the route in km; v2 refers to the real-time average speed through the route, with the unit being km/h;

the calculation process of the predicted travel shortening time of the coordination route is as follows:

assuming that the vehicles uniformly arrive at the intersection before coordination, and the vehicles pass through the intersection without stopping after coordination; the predicted travel shortening time of each road section is the difference value between the travel time before coordination and the travel time after coordination;

t_r＝t_{fruit of Chinese wolfberry}-t_{Coordination device}

Wherein, t_rIs the predicted travel shortening time in seconds; t is t_{Fruit of Chinese wolfberry}Is to pass before coordinationTravel time of the route in seconds; t is t_{Coordination device}Is the travel time of the coordinated passing route, and the unit is second; finally, the predicted travel shortening time is displayed, and the predicted travel shortening times of all the road sections need to be added and summed.

Preferably, the coordinated route and the effect of the coordinated scheme thereof are evaluated, including a recommendation index, the recommendation index of the coordinated route is analyzed according to three indexes of a ratio of the predicted travel shortening time to the route length, a ratio of the predicted parking times to the number of coordinated intersections, and a delay index, and the recommended ratio analysis result of the coordinated route is processed according to quartile points (0, 0.25,0.5, 0.75).

Preferably, the coordination route and the effect of the coordination scheme thereof are evaluated, and the coordination route and the effect of the coordination scheme thereof comprise a matching degree which comprises the following steps:

(I) coordinating the matching degree of the recommended phase and the corresponding phase of the intersection with large flow: comparing the coordination phase recommended by the algorithm with the intersection large-flow phase, taking the times of consistent phases as numerators, and taking the total times of comparison as denominators to obtain the matching degree of the recommended coordination phase and the intersection large-flow corresponding phase;

(II) coordinating the matching degree of the recommended phase difference and the road section travel time: for a section of road L_i，jThe matching degree M of the subsection judgment index expression and the coordinated recommended phase difference of the road section and the travel time of the road section_otIs calculated as follows:

wherein M is_otRecommending the matching degree of the phase difference and the road section travel time for coordination; k is a radical of_tRecommending a time period for the coordination;

is a section of road L_i，jThe number of recommended solutions;

is a section of road L_i，jLength of (km);

is k_tSection L of a time segment_i，jUniform velocity (km/h); o is_j(k_t) Recommended section of road L for time period_i，jThe phase difference of (a);

(III) matching degree of the coordination recommendation period and the total flow of the coordination route: for the coordination route j, the subsection judgment index expression of the route and the matching degree M of the coordination recommendation period and the total flow of the coordination route_cvIs calculated as follows:

in the formula, M_cvMatching degree of the coordination recommendation period and the total flow of the coordination route; k is a radical of_tRecommending a time period for the coordination; n is the total number of the coordinated recommendation schemes of the road section; f. of_sum(k_t) Flow sum (vehicle) for the coordinated route in the coordinated recommended time period; c (k)_t) A recommendation period in a coordinated recommendation time period;

(IV) coordinating the matching degree of the recommended green light time and the flow of the coordination scheme: for intersection I_iThe matching degree M of the segmentation judgment index expression and the coordinated recommended green light time of the intersection and the total flow of the coordinated route_gvThe calculation is as follows:

wherein M is_gvRecommending the matching degree of the green light time and the total flow of the coordination route for coordination; k is a radical of_tRecommending a time period for the coordination; n is_iIs a road junction I_iTotal number of recommended solutions;

is a road junction I_iFlow (vehicle) at a coordination phase of a coordination recommendation period;

is a road junction I_iAt k_tGreen time of the coordinated phase of the time segments.

A recommendation system for a coordination route and a coordination scheme of urban traffic comprises a data module, an algorithm module, a coordination index calculation module, a coordination effect evaluation module and an interaction visualization module;

the data module is used for collecting and preprocessing traffic data;

the algorithm module obtains a recommended coordination route through a threshold judgment method, selects a target road section for the determined coordination route, and establishes a state equation of the change of the flow relation of the target road section along with time; extracting state parameters, control parameters, state time-varying parameters and control time-varying parameters in a state equation of the target road section, constructing an optimal control system of the target road section based on a DDALQR algorithm, and solving optimal time-varying parameters of the target road section, wherein the optimal time-varying parameters describe optimal current-time control parameters under current-time state parameters; calculating the coordinated green light time of the coordinated intersections at the two ends of the target road section at the current moment according to the optimal time-varying parameter of the target road section and the state of the target road section at the current moment;

the calculation coordination index module is used for calculating a coordination time interval, a coordination phase difference, a coordination cycle and a coordination scheme of a coordination route;

the coordination effect evaluation module is used for evaluating the coordination route and the effect of the coordination scheme thereof to obtain the delay index, the predicted travel shortening time and the coordination recommendation index of the coordination route and the matching degree of the coordination scheme;

the interactive visualization module is used for displaying, and comprises a large screen display and each module page; the module displays the whole or regional road network on a large screen, can click to check the details of the intersection, and can dynamically update the speed of the road section, the efficiency index and the predicted travel shortening time; after the coordination route number on the left side is clicked, a coordination time period and a recommended index star level are given, the position of the coordination route is displayed on the map of the main interface on the right side, and the total travel, the speed, the efficiency index and the predicted travel shortening time of the coordination route are displayed under the large screen.

The invention has the beneficial effects that: according to the method, the phase difference of the signal lamps is automatically set through coordination control under the condition of ensuring the traffic capacity of the branch road according to the traffic state data, so that the stop time of the vehicle at each intersection is reduced, the time for waiting for the red light is minimum when the vehicle reaches each intersection, particularly, the vehicle on a trunk line can smoothly run, the urban traffic operation efficiency is improved, the participation of personnel is reduced, and the coordinated route of the intersection which can be coordinated is automatically recommended. The output result of the invention can be used for the evaluation of the coordinated route recommendation, and can also be used for the output of the coordinated route recommendation scheme and the implementation of the actual urban traffic road section. By carrying out real-time coordination control on the urban traffic trunk line, the urban traffic flow can smoothly flow, and the modernization construction and the improvement of the urban comprehensive traffic control capacity are promoted.

Drawings

FIG. 1 is a system framework diagram of the present invention;

FIG. 2 is a schematic flow chart of a threshold determination method of the algorithm module of the present invention;

FIG. 3 is a schematic diagram of a target road section of the DDALQR algorithm;

FIG. 4 is a schematic diagram of a main road of a south mountain road (river road-Wan Song mountain road) according to an embodiment of the present invention;

FIG. 5 is a graph comparing trends of coordinated recommended coordinated phase flow and coordinated phase difference according to an embodiment of the invention;

FIG. 6 is a graph comparing a coordinated recommendation cycle with a trend of total traffic for a route according to an embodiment of the present invention;

FIG. 7 is a graph comparing a coordinated recommended green time to a coordinated solution flow (Nanshan road Qing Fang) according to an embodiment of the present invention;

FIG. 8 is a graph comparing coordinated recommended green time and coordinated solution traffic for an embodiment of the invention (south mountain road, Vanchong mountain road);

fig. 9 is a time-distance graph of north-south green wave coordination of the south mountain routing according to the embodiment of the present invention.

Detailed Description

The invention will be further described with reference to specific examples, but the scope of the invention is not limited thereto:

example (b): as shown in fig. 1, a recommendation system for a coordination route and a coordination scheme of urban traffic is composed of a data module, an algorithm module, a coordination index calculation module, a coordination effect evaluation module and an interaction visualization module. The method comprises the following specific steps:

the data module is used for data acquisition and data preprocessing: the method comprises the steps of firstly, collecting road network multi-source heterogeneous traffic data from a data acquisition module, and obtaining traffic flow ratio data through the ratio of the flow of each annunciator intersection or road section in each direction to the total flow of the intersection. And selecting data with the flow ratio data in the range (10,90), namely excluding data with the inlet passage flow ratio close to 100% and lower than 10%, and judging the intersection with the abnormal detector data when the flow ratio is greater than 90% or less than 10%.

The algorithm module is used for obtaining a coordination route, selecting a target road section, extracting state parameters, control parameters, state time-varying parameters and control time-varying parameters in a state equation of the target road section, constructing an optimal control system of the target road section based on a DDALQR algorithm, and solving the optimal time-varying parameters of the target road section, wherein the optimal time-varying parameters describe the optimal current-time control parameters under the current-time state parameters. Finally, the coordinated green time is calculated.

1) Acquiring a coordination route: and recommending a coordination route based on the determined trunk road, namely determining that the physical relationship of the coordination intersection is straight or left or right. The intersection of the physical main road is divided into a plurality of groups of coordination routes based on a threshold judgment method according to conditions of settable coordination functions due to different road section distances, saturation data and flow data. The coordination route direction includes unidirectional coordination and bidirectional coordination. According to the actual configuration condition of each intersection of the city, the coordination route recommended by the recommendation algorithm is a one-way coordination route. The coordination route comprises two or more continuous coordination intersections, a road section between two adjacent continuous coordination intersections and a coordination direction between two adjacent continuous intersections. The threshold determination method is shown in fig. 2, and specifically includes the following steps:

step 1: acquiring initial traffic proportion data of a week from a database, setting a traffic proportion threshold condition and a distance threshold condition, and acquiring upstream and downstream relation data of a road section and the maximum value and the minimum value of coordinated intersection data;

step 2: extracting all intersections or road sections meeting the threshold condition, and then judging all data meeting the condition;

and step 3: judging whether the downstream crossing of the crossing meeting the conditions meets the threshold condition and the maximum and minimum limit conditions of the coordinated crossing, adding the downstream crossing meeting the conditions to the crossing coordinated route crossing list, and discarding the coordinated crossing list if the downstream crossing does not meet the conditions;

specifically, the method comprises the following steps: whether the ratio of each direction flow of the intersection to the total intersection amount is greater than a designed threshold value or not: less than a threshold, the direction is discarded; when the flow rate of the downstream crossing in the direction is greater than the threshold value, calculating whether the flow rate of the downstream crossing in the direction is greater than a design threshold value: less than a threshold, the direction is discarded; when the flow rate of the downstream crossing in the direction is greater than the threshold value, calculating whether the flow rate of the downstream crossing in the direction is greater than a design threshold value: less than discarding the intersection, the coordination route cannot be recommended; if the value is larger than the threshold value, the coordination route can be recommended but needs to be continuously judged: if the condition is not met, recommending a 3-intersection coordination route, and if the condition is met: and continuing judging until the condition is not met or the number of the coordinated intersections is more than 6.

And finally, screening the coordinated intersections, and if the intersection distance is more than 800m, the coordination can not be set, segmenting the main road by the scheme, wherein the segmenting step is as follows:

step 1: finding out all intersection (road section) data meeting the conditions;

step 2: grouping the data meeting the conditions according to time and intersections, taking out the upstream intersections and the intersections of the data meeting the conditions, and constructing a large list (all coordinated intersection lists), wherein the content of the list refers to a two-intersection list capable of recommending coordination, such as: and [ [ intersection 1, intersection 2], [ intersection 3, intersection 5] ], judging whether the downstream intersection of each group of data meets a threshold condition or not and whether the data of the coordinated intersection list is less than the maximum value or not, and if so, judging whether the conditions are met. And adding the intersection into the coordinated two-way intersection list, updating all coordinated intersection lists, wherein all the coordinated intersection lists comprise the coordinated two-way intersection list and the coordinated three-way intersection list, and circulating until a threshold condition is met. And finishing the calculation of the coordination intersection of the coordination route.

And 4, step 4: storing the calculation result in a database;

and 5: acquiring the calculated coordination route information;

step 6: setting new flow ratio data or a coordination route under the condition of threshold value condition, and calling a function to return new data.

Examples are: i represents the number of the intersection, I represents the intersection, I_iThe intersection with the number I is shown, and the initial coordination intersection of one coordination route is shown as I_iAccording to the coordination direction, the next intersection is represented as I_i+1In 1 with_i+rRepresenting the R-th intersection according to the coordination direction from the initial coordination intersection, and taking the value of R as [1, R]R +1 is the number of intersections of the coordination route; j denotes a coordination route number, L_i，jIntersection I in coordinated route with representation number j_iAnd junction I_i+1The section between, L_i+1,jIntersection I in coordinated route with representation number j_i+1And junction I_i+2The section between, L_i，j+dThe representation can be from intersection I_iA d-th section of the incoming traffic;

x represents the flow state, x_i，jIs represented by a road section L_i，jFlow state of (x)_i，j+dRepresenting a section of road L_i，j+dD takes on the value [1, D ]]。

As shown in the attached figure 3 of the drawings,one coordination route includes: road junction I_iIntersection I_i+1Intersection I_i+2Three continuous intersections, I_iAnd junction I_i+1The section of road between L_i，jIntersection I_i+1And junction I_i+2The section of road between L_i+1，jThe coordination direction is sequentially the intersection I_iIntersection I_i+1Intersection I_i+2. The intersection Ii is a four-branch intersection and can be taken from the intersection I_iThe number of the road sections flowing into the traffic flow is 3, and the number of the road sections is L_i，j+1、L_i，j+2、L_i，j+3。

2) Selecting a target road section: selecting a target road section from the coordination route, and establishing a state equation of which the flow relation of the inflow and outflow target road sections changes along with time;

and the flow of the target road section at the next moment is equal to the sum of the flow of the target road section at the current moment, the flow of each inflow target road section at the current moment and the flow of each outflow target road section at the current moment.

Referring to FIG. 3, a target link L is selected from the coordinated route numbered j_i，jTarget road section L_i，jThe equation of state of (c):

inflow target section L_i，jThe flow rates of (a) come from: road section L_i，j+1Left-turn lane, road section L_i，j+2Straight lane, road section L_i，j+3Right-turn lane of (1), target road section L at next moment_i，jFlow rate x of_i，j(k +1) is the current flow x_i，j(k) Adding the flow rate flowing into the road section at the current moment, and subtracting the flow rate flowing out of the road section at the current moment.

x_i，j(k+1)＝x_i，j(k)+S_i，j+1(k)Δg_i，j+1(k)+S_i，j+2(k)Δg_i，j+2(k)+S_i，j+3(k)Δg_i，j+3(k) (1)

-S_i，j(k)Δg_i，j(k)

Wherein k represents the kth time point after polymerization for 1 hour; s_i，j+1(k)，S_i，j+2(k)，S_i，j+3(k)，S_i，j(k) Respectively a road section L_i，j+1Left-turn lane ofRoad section L_i，j+2Straight lane, road section L_i，j+3The saturation flow rate of the right-turn lane and the target road section at the moment k is the ratio of the maximum value of the historical flow rate to the green light time;

Δg_i，j+1(k)，Δg_i，j+2(k)，Δg_i，j+3(k)，Δg_i，j(k) respectively a road section L_i，j+1Left-turn lane, road section L_i，j+2Straight lane, road section L_i，j+3The right-turn lane and the green light change time of the target road section at the time k.

3) Extracting state parameters, control parameters, state time-varying parameters and control time-varying parameters in a state equation of the target road section, constructing an optimal control system of the target road section based on a DDALQR algorithm, and solving the optimal time-varying parameters of the target road section, wherein the optimal time-varying parameters describe the optimal current-time control parameters under the current-time state parameters.

a) Extracting state parameters, control parameters, state time-varying parameters and control time-varying parameters in a state equation of the target road section;

the target link state equation (1) can be converted into:

the state parameter is x_i，j(k+1)、x_i，j(k)；

The control parameter is

The state time varying parameter is 1;

controlling a time-varying parameter S_i，j+1(k) S_i，j+2(k) S_i，j+3(k) -S_i，j(k)]。

b) Constructing an optimal control system of a target road section based on a DDALQR algorithm:

the state equation of the target road section can be converted into a state equation in a DDALQR algorithm:

x(k+1)＝A(k)x(k)+B(k)Δg(k) (3)

wherein x (k +1) and x (k) are the flow of the target road section at the moment k +1 and the moment k;

a (k) is a state time-varying parameter at time k;

b (k) is a time-varying control parameter at time k.

Constructing A (k), and obtaining a time-varying parameter A (k) as an identity matrix of 1 order:

A(k)＝1 (4)

so time varying parameter B_i，j+4(k)：B_i，j+4(k)＝[S_i，j+1(k) S_i，j+2(k) S_i，j+3(k) -S_i，j+4(k)] (5)

The calculation formula of the control performance index J of the optimal control system is as follows:

wherein the content of the first and second substances,

is the optimal control quantity;

q (k) is a traffic state weighting matrix at the moment k;

and R (k) is a traffic state control weighting matrix at the moment k.

The optimal time-varying parameter l (k) that satisfies the performance index J satisfies the following relationship:

wherein the content of the first and second substances,

the phase-based green duration.

c) Obtaining an optimal time-varying parameter of the target road section, wherein the optimal time-varying parameter describes an optimal current-time control parameter under a current-time state parameter;

by determining optimum control quantities

The control performance index J in equation (6) is minimized because both terms of J are non-negative and minimizing J minimizes each of its terms.

Next, for the link L_i，jTo minimize the first term in the control performance index of equation (6), the first term is first transformed into:

wherein each term of the structure Q (x) is

The smallest q (x) means that the denominator of q (x) is the largest, i.e. the road capacity is larger, i.e. the load between the intersection and the road section is balanced as much as possible.

Therefore, further elaboration may yield the expression of the time-varying parameter q (x) as:

in the formula: and C (k) is the time-varying road section bearing capacity, namely the road traffic capacity, and the value is represented by the historical maximum flow.

For road section L, the same principle applies_i，jIn order to minimize the second term in the control performance index of equation (6), the second term is first used

The transformation is:

wherein the content of the first and second substances,each term of our structure R (x) is

To minimize R (x), the numerator of R (x) is minimized, i.e., the green time rate of change amplitude of the road is minimized.

Therefore, the time-varying parameter r (k) can be obtained through sorting:

because of the nonlinear characteristic and the time-varying characteristic of a traffic system, in order to simplify calculation, only the lanes on a coordinated route are considered, when a plurality of lanes are involved, the average flow rates of the lanes are added, when a road section or the lanes correspond to a plurality of phases, parameters A (k), B (k), Q (k) and R (k) are obtained through equations (4), (5), (10) and (12), an equation (8) is taken as an objective function, equations (2), (7) and (6) are taken as constraint conditions, and a time-varying parameter L (k) can be calculated by solving the optimization problem;

d) the DDALQR algorithm is further introduced:

the DDALQR algorithm is based on the following equation of state:

x(k+1)＝A(k)x(k)+B(k)Δg(k) (13)

wherein x (k +1) is a state vector of the number of vehicles released at the green light time of each entrance road section at the moment of k + 1; x (k) is a state vector of the number of vehicles released at the green time of each entrance section at the time k; Δ g (k) is a control quantity of the green light time at the moment k; a (k) is a parameter quantity that varies with time; b (k) is a parameter that varies with time.

Meanwhile, the specific expression of Δ g (k) is as follows:

Δg(k)＝g(k)-g^N (14)

wherein, g^NA phase-based green time duration matrix on an entrance section of the intersection; g (k) is a green time duration matrix on the entrance section of the intersection at the current k moment.

According to the optimal control theory, the optimal solution of the control input quantity is

Δg^*(k)＝-R(k)^-1B^T+P^*x(k) (15)

In the formula, P^*Is an N-th order positive solution of the discrete algebraic Riccati equation Δ g (k). The equation (15) may be changed to:

according to the feedback control theory and the DDALQR algorithm, the specific expression form of g (k) can be obtained as follows:

g(k)＝g^N-L(k)x(k) (17)

wherein l (k) is a time-varying p × q matrix, p is the number of upstream channels on a road section, and q is the number of road sections on an entrance road; according to the control theory, the calculation formula of the control performance index J is as follows:

wherein Q (k) is a state weighting matrix, which represents the relative importance of each state error and is a positive definite matrix; r (k) is a control weight matrix, representing the relative importance of input energy consumption, which is a positive definite matrix.

In combination with the actual traffic problem, when the DDALQR algorithm is adopted to carry out the split control on the urban traffic, the urban traffic system can be simplified into a discrete time system with infinite time length, and the formula (18) is simplified, so that the calculation formula for controlling the performance index can be discretized into the following formula:

wherein x is_k(k) The state vector of the number of vehicles released at the green light time of each entrance road section at the moment k; Δ g_k(k) The control quantity of the green light time at the moment k;

is the optimal control quantity; q (k) is traffic state at time kA state weighting matrix; and R (k) is a traffic state control weighting matrix at the moment k.

4) Calculating the coordinated green light time: calculating the coordinated green light time of the coordinated intersections at the two ends of the target road section at the current moment according to the optimal time-varying parameter of the target road section and the state of the target road section at the current moment;

the optimal timing scheme calculated by each single intersection according to the Webster timing method is used as an initialization scheme. In the actual coordination direction, there may be a case where there is coordination between a plurality of intersections close in distance, and if the intersections Ii and I are close to each other_i+2Intersection I_i+3There is coordination, as shown in fig. 3, which is calculated as follows:

examples are: step 1: according to the coordination direction, calculating the intersection I by using the Webster method_iThe optimal green light time scheme is used as an initialization green light time scheme;

step 2: calculation of coordinated intersection I by using DDALQR algorithm_i+1A coordinated green time scheme of (1);

step 3: updating the intersection I according to the calculation result of the step2_i+1The green light time scheme distributes the updated scheme in proportion as the intersection I_i+1The optimal green light time coordination scheme is adopted;

step 4: using DDALQR algorithm and updated intersection I_i+1Intersection I coordinated with green light time scheme calculation_i+2The optimal coordinated green time scheme.

Therefore, along the coordination direction of the coordination route, each coordination intersection performs the above-described operation until the target link to the last coordination intersection is operated.

Optimization of

And acquiring the flow x (k) of the target road section at the current moment, acquiring an optimal control vector according to the optimal time-varying parameter of the target road section, and finishing the split horizon control of the DDALQR algorithm.

The calculation coordination index module is used for calculating a coordination time interval, a coordination phase difference, a coordination period and a coordination scheme of a coordination route; the method specifically comprises the following steps:

1) calculating a coordination period; and calculating each coordination period of the coordination route according to the saturation data of the coordination intersection based on the determined coordination route. Because the aim of coordination control needs to solve the most outstanding traffic contradiction problem on the road section in priority, the integral saturation of the intersection in each direction in the traffic time period is taken as reference data, and a control strategy time-sharing scheme is made for each time period all day, so that the coordination time period is calculated. The overall non-highest saturation of the intersections in all directions in the traffic time period is taken as a reference basis, but the calculation finds that a large amount of resources are consumed and the actual traffic control is not greatly influenced, so that the calculation finds that the overall highest saturation of the intersections in all directions in the traffic time period is taken as reference data, a control strategy time-sharing scheme is made for all time periods in the whole day, the efficiency of the coordination time period is calculated to be the highest, and the index is adopted for calculation.

2) Calculating a coordination phase: and calculating all phases which can be coordinated according to the corresponding relation between the corresponding road section of the coordinated intersection and the lamp group and the corresponding relation between the lamp group and the phase, namely the coordinated phase. According to the actual traffic condition of Hangzhou, right turn coordination is eliminated, and multiple periods of adjustment are needed when the coordination scheme is set to the scheme. In the same time period, a single intersection can only appear in one group of coordination routes, and only one group of coordination routes with the largest number of intersections can be reserved in the same time period of coordination routes.

3) Calculating a coordination route phase difference: the green signal ratio control in the algorithm module of the invention also needs periodic control, because all periodic flow control methods are longer in period and reduce the proportion of lost time, thereby increasing the traffic capacity of the intersection. On the other hand, if the traffic flow is low, the period is long, which leads to the waste of green light time, and is not beneficial to the reasonable allocation of urban traffic resources.

The function realizes the calculation of the I signal period of the intersection, and the calculation formula is as follows

Wherein l_IDistance between the upstream intersection and the downstream intersection; v. of_IThe average speed of the vehicle between the upstream intersection and the downstream intersection;

the value of the control parameter is continuously calculated to obtain an optimal result; v. of^cIs constant, according to the actual traffic equipment data, v in the invention^cTaking a value of 15 km/h; x is the number of_L(k) The traffic flow in the 1h period on the road section L;

is the maximum traffic flow over the historical period on the road segment L.

4) Calculating a coordination route period: the invention realizes the adjustment of the intersection signal period through a feedback algorithm. The algorithm adjusts the signal cycle time of the intersection to adapt the intersection cycle to the maximum saturation of the intersection in the road network. The periodic updating method of the intersection comprises the following steps:

for intersection I, the calculation formula of the intersection signal period is as follows:

C(k)＝C^N+K^c(σ(k)-σ^N) (23)

σ_I(k)＝x_L/x_L ^max (24)

wherein C (k) is the intersection signal period at the moment k; c^NThe period duration of the existing signal; k^CThe variable control parameters have values that affect the intensity of traffic control, and the values are averaged according to the routes that are calculated multiple times and meet the coordination requirement. Sigma_I(k) The current load of the intersection; x is the number of_LThe traffic flow in the period of the road section L;

the maximum traffic flow in the road section period is obtained; sigma^NThe average load of the coordinated route is the average value of the loads of all intersections under the coordinated route.

5) And (3) calculating a coordination scheme: and selecting a target road section according to the coordination direction of the coordination route, and calculating a coordination scheme of the coordination route. And allocating the coordinated green light time calculated by the algorithm module to each phase according to the original phase allocation proportion. As shown in figure 3, if the intersection I_iIntersection I_i+1Coordination exists, and the calculation process of the coordination scheme is as follows:

along the coordinated direction, for intersection I_iIntersection I_i+1Target link L of_i，jObtaining an optimal control vector Δ g_i，j+1(k)、Δg_i，j+2(k)、Δg_i，j+3(k)、Δg_i，j(k) In that respect Assumed intersection I_iThe initialization phases calculated by the Webster method are distributed as an A phase 30s, a B phase 30s and a C phase 30s, and the period of the intersection Ii is 90 s. The calculated coordination period is 180s, and when the coordination direction is the following intersection I_iTo the intersection I_i+1If the coordination phase is the a phase, and the green time of the coordination phase a is calculated as 70s according to the formula (17), and the remaining 90s are respectively allocated to the other phases in proportion to the original B phase and C phase, the B phase is 45s, and the C phase is 45 s. Has finished the following road junction I_i，jTo the intersection I_i+1，jGreen ratio control and periodic control of the coordinated route.

The coordination effect evaluation module is used for evaluating the coordination route and the effect of the coordination scheme thereof to obtain the delay index of the coordination route, the predicted travel shortening time, the coordination recommendation index and the matching degree of the coordination scheme; the method comprises the following specific steps:

1) delay index: the delay index of the road refers to the ratio of the real-time passing time of the passing route to the passing time of the passing route in a clear state. Calculated according to the following formula:

d＝t₁/t₂ (25)

wherein d is a delay index of the road; t is t₁Real-time road passing time(s) through a route; t is t₂Refers to the time(s) of passage of the route in the clear state; wherein, the real-time transit time of the passing route refers to the ratio of the distance of the passing route to the designed speed of the route under the condition of free flow, namely according toThe following formula calculates:

t₁＝∑s/∑v₁ (26)

wherein, t₁Real-time road passing time(s) through a route; s is the distance through the route (km); v. of₁Refers to the designed speed (km/h) through the route; under the condition of free flow, the designed speeds of all the sections of the route are different, and then the sections of different sections on the route are summed up during subsection calculation. The real-time road passing time of the passing route refers to the ratio of the distance of the passing route to the design speed of the route in a smooth state, namely the real-time road passing time is calculated according to the following formula:

t₂＝∑s/∑v₂ (27)

wherein, t₂Refers to the time(s) of passage through the route in the clear state; s is the distance through the route (km); v2 refers to the real-time average speed (km/h) through the route;

2) predicted stroke shortening time: and assuming that the vehicles uniformly arrive at the intersection before coordination, and the vehicles pass through the intersection without stopping after coordination. The predicted travel shortening time for each link is the difference between the pre-coordination travel time and the post-coordination travel time. Namely, it is

t_r＝t_{Fruit of Chinese wolfberry}-t_{Coordination device} (28)

Wherein, t_rIs the predicted travel reduction time(s); t is t_{Fruit of Chinese wolfberry}Is the travel time(s) of the pre-coordination transit route; t is t_{Coordination device}Is the travel time(s) through the route after coordination; finally, the predicted travel shortening time is displayed, and the predicted travel shortening times of all the road sections need to be added and summed.

3) Recommendation index: the recommendation index of the invention is analyzed according to three indexes of the ratio of the predicted travel shortening time to the route length, the ratio of the predicted stopping times to the number of coordinated intersections and the delay index, and is processed according to quartile points (0, 0.25,0.5 and 0.75) according to the analysis result of the ratio of the recommended coordinated routes.

a) Predicted trip reduction time/route length:

the four-star level, the interval belongs to [0.75,1], and the coordination effect is best in the interval;

three stars — interval belongs to [0.5,0.75 ];

two stars — interval belongs to [0.25,0.5 ];

one star level- -interval belongs to [0,0.25 ];

b) the ratio of the estimated number of parking times to the number of coordinated intersections:

four stars-interval belongs to [0,0.25], and the coordination effect is best in the interval;

three stars — interval belongs to [0.25,0.5 ];

two stars — interval belongs to [0.5,0.75 ];

one star level- -the interval belongs to [0.75,1 ].

c) Delay index:

four stars-interval belongs to [0,0.25], and the coordination effect is best in the interval;

three stars — interval belongs to [0.25,0.5 ];

two stars — interval belongs to [0.5,0.75 ];

one star level- -the interval belongs to [0.75,1 ].

The total score takes the same rank first, and no same rank takes the highest value of the three indices.

4) Matching degree: the matching degree of the coordination scheme is that the expected variation trends of the period and the duration of the green light are compared with the variation trend of relevant indexes such as flow and the like, and the matching degree is obtained by dividing the expected comparison times by the total comparison times. The value range of the matching degree is [0,1 ]. The higher the degree of matching, the higher the accuracy of the decision algorithm recommendation.

The interaction visualization module comprises a large screen display and each module page. The module displays the whole or regional road network on a large screen, can click to check the details of the intersection, and can dynamically update the speed of the road section, the efficiency index and the predicted travel shortening time. After the coordination route number on the left side is clicked, a coordination time period and a recommended index star level are given, the position of the coordination route is displayed on the map of the main interface on the right side, and the total travel, the speed, the efficiency index and the predicted travel shortening time of the coordination route are displayed under the large screen.

A method for recommending a coordinated route and a coordinated plan for urban traffic, comprising the steps of:

(1) collecting multi-source heterogeneous traffic data of a road network, and preprocessing the data;

(2) acquiring a coordination route based on a threshold judgment method, selecting a target road section from the coordination route, and establishing a state equation of the flow relation of the target road section changing along with time;

(3) extracting state parameters, control parameters, state time-varying parameters and control time-varying parameters in a state equation of the target road section, constructing an optimal control system of the target road section based on a DDALQR algorithm, and solving optimal time-varying parameters of the target road section, wherein the optimal time-varying parameters describe optimal current-time control parameters under current-time state parameters;

(4) calculating the coordinated green light time of the coordinated intersections at the two ends of the target road section at the current moment according to the optimal time-varying parameter of the target road section and the state of the target road section at the current moment;

(5) respectively calculating a coordination period, a coordination phase difference, a coordination cycle and a coordination scheme of the coordination route based on the coordination route;

(6) evaluating the coordination route and the effect of the coordination scheme thereof to obtain a delay index, a predicted travel shortening time and a coordination recommendation index of the coordination route and the matching degree of the coordination scheme; and finally, outputting and displaying the evaluation result, the coordination route and the coordination scheme together for the reference of a user.

In this embodiment, a recommended coordination route of a south mountain road (a river road-a wangsu road) in the state of hangzhou is taken as an example for explanation, the length of the road coordination route is 729 meters, and coordination intersections include a south mountain road-a river road, a south mountain road-a qing wave street, a south mountain road-a wangsu road, and the route has no pedestrian crossing or interference factors along the route. And selecting the recommended one-week coordination scheme as an evaluation object. The method specifically comprises the following steps:

1. arterial profile of the coordination route:

first, the main road of the recommended coordinated route is analyzed, and the details of the intersection of the main road are shown in fig. 4. The coordination route of the algorithm output of the present invention is shown in table 1 below:

TABLE 1

2. And (3) coordinating and recommending intersection analysis:

considering two indexes of intersection distance and intersection canalization, after comparing and screening each intersection, calculating the coordination intersections including south mountain roads-river streets, south mountain roads-Qing Bo street, south mountain roads-Wan Song mountain roads into a group of coordination control routes.

3. And (3) coordinating and recommending period analysis:

the road saturation is one of important indexes reflecting the road service level, and the coordination time interval is judged according to the saturation threshold of the coordination road section. The calculation formula of the saturation is the ratio of the maximum traffic volume to the maximum traffic capacity. Namely, it is

dt＝V_max/C_max (29)

Wherein dt is the saturation; v_maxIs the maximum traffic volume; c_maxIs the maximum traffic capacity; respectively carrying out time interval analysis on the saturation of the three key intersections, combining factors such as actual coordination transition time and the like, and coordinating the maximum value of the saturation of the intersection within a certain time>0.8 or intersection saturation minimum<0.2, not recommending the coordination of the time interval, and judging that the coordination time interval is 8:00:00-9:00: 00.

4. And (3) coordinating and recommending direction analysis:

in combination with the actual traffic condition of the urban road, when the ratio of the flow of the inlet road at the coordination intersection is less than 0.2 or greater than 0.9, the coordination of the direction is not recommended, and the coordination direction of the sample is from north to south.

5. And coordinating the matching degree of the recommended scheme and the actual:

the method comprises the steps of respectively evaluating the matching degree of the green light duration of a recommended coordination phase, a coordination phase difference, a coordination period and a coordination phase with the actual situation of a route, wherein the specific evaluation contents comprise the matching degree of the recommended coordination phase and a phase corresponding to a large flow, the matching degree of the recommended phase difference and a road section travel time, the matching degree of the recommended period and a total flow and the matching degree of the green light duration of the recommended coordination phase and a phase flow.

1) Coordinating matching degree of recommended phase and intersection large-flow corresponding phase

Coordinating the matching degree of the recommended phase and the corresponding phase of the intersection with large flow: and comparing the coordination phase recommended by the algorithm with the intersection large-flow phase, taking the times of consistent phases as numerators, and taking the total times of comparison as denominators to obtain the matching degree of the recommended coordination phase and the intersection large-flow corresponding phase.

According to calculation, the matching degree of the recommended coordination phase of the evaluation selection scheme and the corresponding phase of the intersection high flow rate is 1. Table 2 is a table of matching recommended coordination phases with total traffic at the intersection. As can be seen from the table, the matching degree of the coordination phase of the clear street in the south mountain is 1, and the matching degree of the coordination phase of the wangsu mountain road in the south mountain is 1.

Road section	Wave clearing street on south mountain road	Wan Song Ling road on south mountain road
			Coordinating phase matching	1	1

TABLE 2

2) Coordinating the matching degree of the recommended phase difference and the road section travel time: as the travel time between the two portals increases (decreases), the algorithm expects the recommended coordinated phase difference to increase (decrease). And comparing whether the recommended change of the coordination phase difference is consistent with the change trend of the travel time between the two ports. And taking the times with consistent comparison results as numerators, and taking the total times of comparison as denominators to obtain the matching degree of the recommended coordination phase difference and the road section travel time. Here, the link distance divided by the link average speed is used as the travel time of the link.

For a section of road L_i，jThe matching degree M of the subsection judgment index expression and the coordinated recommended phase difference of the road section and the travel time of the road section_otIs calculated as follows:

wherein M is_otRecommending the matching degree of the phase difference and the road section travel time for coordination; k is a radical of_tRecommending a time period for the coordination;

is a section of road L_i，jThe number of recommended solutions;

is a section of road L_i，jLength of (km);

is k_tSection L of a time segment_i，jUniform velocity (km/h); o is_j(k_t) Recommended section of road L for time period_i，jThe phase difference of (1).

Due to the lack of speed data, the matching degree of the coordinated recommended phase difference and the road section travel time is further simplified, the trend of the coordinated phase flow and the coordinated phase difference is compared in the evaluation, and the coordinated phase difference and the coordinated phase flow have opposite trends and are in line with expectation as shown in the attached figure 5.

3) Matching degree of coordination recommendation period and total flow of coordination route

Matching degree of the coordination recommendation period and the total flow of the coordination route: when the total flow of the coordinated route increases (decreases), the algorithm anticipates that the recommended period value increases (decreases). And comparing the variation trend of the total flow of the coordination route with the variation trend of the recommendation period to determine whether the variation trend is consistent with the variation trend of the recommendation period. And taking the times with consistent comparison results as numerators, and taking the total times of comparison as denominators to obtain the matching degree of the recommendation period and the total flow of the intersection.

For the coordination route j, the subsection judgment index expression of the route and the matching degree M of the coordination recommendation period and the total flow of the coordination route_cvIs calculated as follows:

in the formula, M_cvMatching degree of the coordination recommendation period and the total flow of the coordination route; k is a radical of_tRecommending a time period for the coordination; n is the total number of the coordinated recommendation schemes of the road section; f. of_sum(k_t) Flow sum (vehicle) for the coordinated route in the coordinated recommended time period; c (k)_t) Is the recommendation cycle in the coordinated recommendation period.

According to calculation, the matching degree of the recommended period of the evaluation selection scheme and the total flow of the coordination route is 0.754. And (3) coordinating a comparison graph of the recommendation period and the trend of the total flow of the route, and referring to fig. 6, the recommendation period and the trend of the total flow of the route are approximately the same.

4) Matching degree of coordinating recommended green light time and coordinating scheme flow

Coordinating the matching degree of the recommended green light time and the flow of the coordination scheme: for the same intersection, the traffic flow increases, and the algorithm expects that the green duration of the corresponding phase will increase. And comparing the flow rate variation trend with the variation trend of the green light time length of the corresponding phase. And taking the comparison result with the same change trend as a numerator, and taking the total times of comparison as a denominator to obtain the matching degree of the flow and the green light duration of the corresponding coordination phase.

For intersection I_iThe matching degree M of the segmentation judgment index expression and the coordinated recommended green light time of the intersection and the total flow of the coordinated route_gvIs calculated as follows：

Wherein M is_gvRecommending the matching degree of the green light time and the total flow of the coordination route for coordination; k is a radical of_tRecommending a time period for the coordination; n is_iIs a road junction I_iTotal number of recommended solutions;

is a road junction I_iFlow (vehicle) at a coordination phase of a coordination recommendation period;

is a road junction I_iAt k_tGreen time of the coordinated phase of the time segments.

According to calculation, the matching degree of the flow of the selected scheme and the green light duration of the corresponding phase is 0.66. And table 3 is a table of recommended coordination phase green light duration and flow matching. As can be seen from the table, the matching degree of the south mountain clear street is 0.65, and the matching degree of the south mountain Wansong mountain road is 0.67. A comparison graph of green light time recommended by coordination of the clear wave street in the south mountain road and flow of a coordination scheme is shown in an attached figure 7. A comparison graph of green light time recommended by coordination of the Wansong mountain road of the south mountain road and flow rate of a coordination scheme is shown in an attached figure 8.

Road section	Wave clearing street on south mountain road	Wan Song Ling road on south mountain road
			Green light duration matching degree	0.65	0.67

TABLE 3

In summary, based on the above calculation on matching, the matching degree of the recommended coordination phase of the current evaluation selection scheme and the phase corresponding to the large flow is 1, the matching degree of the recommended cycle and the total flow of the route is 0.754, and the matching degree of the recommended coordination phase green light duration and the phase flow is 0.66. And (3) combining the road section condition and the actual signal timing condition of each road junction, drawing a north-south green wave coordination time distance graph of the south mountain route, and referring to fig. 9. According to the time-distance diagram, the direction and specific coordination situation of the south mountain road coordination can be calculated by combining the actual situation (considering branch line interference and base entrance and exit interference). The feasibility and application value of the coordinated recommendation algorithm and the coordinated recommendation scheme of the invention can be seen from the above examples. The invention carries out off-line coordination control on the urban traffic trunk line and promotes the modernized construction of urban comprehensive traffic control capacity.

While the invention has been described in connection with specific embodiments thereof, it will be understood by those skilled in the art that various changes in form and details may be made therein without departing from the spirit and scope of the invention as defined by the appended claims.

Claims (9)

1. A method for recommending a coordinated route and a coordinated plan for urban traffic, characterized by comprising the steps of:

(1) collecting multi-source heterogeneous traffic data of a road network, and preprocessing the data;

(2) acquiring a coordination route, selecting a target road section from the coordination route, and establishing a state equation of the flow relation of the target road section changing along with time;

(3) extracting state parameters, control parameters, state time-varying parameters and control time-varying parameters in a state equation of the target road section, constructing an optimal control system of the target road section based on a DDALQR algorithm, and solving optimal time-varying parameters of the target road section, wherein the optimal time-varying parameters describe optimal current-time control parameters under current-time state parameters; the method specifically comprises the following steps:

(3.1) extracting state parameters, control parameters, state time-varying parameters and control time-varying parameters in the state equation of the target road section;

target road section L_i，jEquation of state (1) of (a) translates to:

then the state parameter is x_i，j(k+1)、x_i，j(k)；

The control parameter is

The state time varying parameter is 1;

controlling a time-varying parameter S_i，j+1(k) S_i，j+2(k) S_i，j+3(k) -S_i，j(k)]；

(3.2) constructing an optimal control system of the target road section based on a DDALQR algorithm;

converting the state equation of the target road section into a state equation in a DDALQR algorithm:

x(k+1)＝A(k)x(k)+B(k)Δg(k) (3)

wherein x (k +1) and x (k) are the flow of the target road section at the moment k +1 and the moment k; a (k) is a state time-varying parameter at time k; b (k) is a control time-varying parameter at time k;

constructing A (k), and obtaining a time-varying parameter A (k) as an identity matrix of 1 order:

A(k)＝1 (4)

so time varying parameter B_i，j+4(k)：

B_i，j+4(k)＝[S_i，j+1(k) S_i，j+2(k) S_i，j+3(k) -S_i，j+4(k)] (5)

The calculation formula of the control performance index J of the optimal control system is as follows:

wherein the content of the first and second substances,

is the optimal control quantity; q (k) is a traffic state weighting matrix at the moment k; r (k) is a traffic state control weighting matrix at the moment k; the optimal time-varying parameter l (k) that satisfies the performance index J satisfies the following relationship:

wherein the content of the first and second substances,

green time on a phase basis;

(3.3) obtaining the optimal time-varying parameter of the target road section, wherein the optimal time-varying parameter describes the optimal current-time control parameter under the current-time state parameter;

first by determining the optimum control quantity

The control performance index J in the formula (6) is enabled to be extremely small;

secondly for the section of road L_i，jIn order to minimize the first term in the control performance index of equation (6), the first term is first used

The transformation is:

wherein each term of the structure Q (x) is

Minimizing Q (x) maximizes the denominator of Q (x); thus, the expression for the time-varying parameter q (x) is obtained as:

in the formula: c (k) represents the bearing capacity of the time-varying road section, and represents the value of C (k) by adopting the historical maximum flow value;

for road section L, the same principle applies_i，jIn order to minimize the second term in the control performance index of equation (6), the second term is first used

The transformation is:

wherein each of the structures R (x) is

Minimizing r (x) minimizes the molecules of r (x), thus yielding the time-varying parameter r (k):

in summary, the parameters a (k), b (k), q (k), and r (k) are obtained through the equations (4), (5), (10), and (12), and the optimal time-varying parameter l (k) can be calculated by solving the optimization problem with the equation (8) as the objective function and the equations (2), (7), and (6) as the constraint conditions;

(4) calculating the coordinated green light time of the coordinated intersections at the two ends of the target road section at the current moment according to the optimal time-varying parameter of the target road section and the state of the target road section at the current moment;

(5) respectively calculating a coordination period, a coordination phase difference, a coordination cycle and a coordination scheme of the coordination route based on the coordination route;

(6) the coordination route and the effect of the coordination scheme thereof are evaluated.

2. The method of claim 1, wherein the method comprises: when a coordination route is obtained, the coordination route is recommended based on the determined trunk road, and the intersection of the trunk road is divided into a plurality of groups of coordination routes according to a threshold judgment method; the coordination route comprises two or more continuous coordination intersections, a road section between two adjacent continuous coordination intersections and a coordination direction between two adjacent continuous intersections; the main road intersection is divided into a plurality of groups of coordination routes according to a threshold judgment method, and the following steps are specifically performed:

(i) acquiring initial flow ratio data of a week from a database, setting a flow ratio threshold condition and a distance threshold condition, and acquiring upstream and downstream relation data of a road section and the maximum value and the minimum value of coordinated intersection data;

(ii) extracting all intersections or road sections meeting the threshold condition, and then judging all data meeting the condition;

(iii) judging whether the downstream crossing of the crossing meeting the conditions meets the threshold condition and the maximum and minimum limit conditions of the coordinated crossing, adding the downstream crossing meeting the conditions to the crossing coordinated road route crossing list, and discarding the coordinated crossing list if the downstream crossing does not meet the conditions; screening the coordinated intersections, segmenting the main road intersections which cannot be coordinated, and storing results to a database;

(iv) acquiring the calculated coordination route information;

(v) setting new flow ratio data or a coordination route under the condition of threshold value condition, and calling a function to return new data.

3. The method of claim 1 for recommending a coordinated route and a coordinated plan for urban traffic,

the method is characterized in that: the establishing of the state equation in the step (2) specifically comprises the following steps: let us choose the target link L from the coordinated route numbered j_i，jEstablishing a target road section L_i，jThe equation of state of (c):

wherein the target section L is flowed into_i，jThe flow rates of (a) come from: road section L_i，j+1Left-turn lane, road section L_i，j+2Straight lane, road section L_i，j+3Right-turn lane of (1), target road section L at next moment_i，jFlow rate x of_i，j(k +1) is the current flow x_i，j(k) Adding the flow flowing into the road section at the current moment, and subtracting the flow flowing out of the road section at the current moment; the equation of state is shown below:

k represents the kth time; s_i，j+1(k)，S_i，j+2(k)，S_i，j+3(k)，S_i，j(k) Respectively a road section L_i，j+1Left-turn lane, road section L_i，j+2Straight lane, road section L_i，j+3The saturation flow rate of the right-turn lane and the target road section at the moment k is the ratio of the maximum value of the historical flow rate to the green light time;

Δg_i，j+1(k)，Δg_i，j+2(k)，Δg_i，j+3(k)，Δg_i，j(k) respectively a road section L_i，j+1Left-turn lane, road section L_i，j+2Straight lane, road section L_i，j+3The right-turn lane and the green light change time of the target road section at the time k.

4. The method of claim 1, wherein the method comprises: the step (4) is specifically as follows:

if the intersection I_iIntersection I_i+2Intersection I_i+3There is a coordination of all the components,the calculation process of the coordinated green time is as follows:

(a) according to the coordinated direction, the junction I is calculated by using the Webster method_iThe optimal green light time scheme is used as an initialization green light time scheme;

(b) calculation of coordinated intersection I by using DDALQR algorithm_i+1A coordinated green time scheme of (1);

(c) updating the intersection I according to the calculation result of the step (b)_i+1The green light time scheme distributes the updated scheme in proportion as the intersection I_i+1The optimal green light time coordination scheme is adopted;

(d) based on DDALQR algorithm, updated intersection I is utilized_i+1Intersection I coordinated with green light time scheme calculation_i+2The optimal green light time coordination scheme is adopted; along the coordination direction of the coordination route, each coordination intersection carries out the operation until the target road section of the last coordination intersection is operated; and acquiring the flow x (k) of the target road section at the current moment, acquiring an optimal control vector according to the optimal time-varying parameter of the target road section, and finishing the split horizon control of the DDALQR algorithm.

5. The method of claim 1, wherein the method comprises: the step (5) is specifically as follows:

(5.1) calculating the coordination period: calculating each coordination period of the coordination route according to saturation data of the coordination intersection based on the determined coordination route;

(5.2) calculating a coordination phase: calculating all phases which are possible to coordinate according to the corresponding relation between the corresponding road section of the coordinated intersection and the lamp group and the corresponding relation between the lamp group and the phase, namely the coordinate phase; in the same time period, a single intersection can only appear in one group of coordination routes, and the coordination routes in the same time period can only reserve one group of coordination routes with the largest number of intersections;

(5.3) calculating the coordination route phase difference: for the calculation of the signal period of the intersection I, the calculation formula is as follows:

wherein l_IDistance between the upstream intersection and the downstream intersection; v. of_IThe average speed of the vehicle between the upstream intersection and the downstream intersection;

the value of the control parameter is continuously calculated to obtain an optimal result; v. of^cIs constant and is determined according to the actual traffic equipment data;

x_L(k) the traffic flow in the 1h period on the road section L;

the maximum traffic flow in the historical period on the road section L;

(5.4) calculating a coordination period: adjusting the intersection signal period through a feedback algorithm; the algorithm adjusts the signal cycle time of the intersection to enable the intersection cycle to adapt to the maximum saturation of the intersection in the road network, wherein the intersection cycle updating method comprises the following steps:

for intersection I, the calculation formula of the intersection signal period is as follows:

C(k)＝C^N+x^C(σ(k)-σ^N)

σ_I(k)＝x_L/x_L ^max

wherein C (k) is the intersection signal period at the moment k; c^NThe period duration of the existing signal; k^cThe variable control parameters have values which influence the intensity of traffic control, and the values are averaged in the routes which meet the coordination requirement according to multiple calculations; sigma_I(k) The current load of the intersection; x is the number of_LThe traffic flow in the period of the road section L;

the maximum traffic flow in the road section period is obtained;σ^Nthe average load of the coordinated route is the average value of the loads of all intersections under the coordinated route;

(5.5) calculating a coordination scheme: selecting a target road section according to the coordination direction of the coordination route, and calculating a coordination scheme of the coordination route; and (4) distributing the coordinated green light time obtained by the calculation in the step (4) to respective phases according to the original phase distribution proportion, thereby completing the green signal ratio control and the period control of the coordinated route.

6. The method of claim 1, wherein the method comprises: evaluating the effects of the coordination route and the coordination scheme thereof, wherein the effects comprise a delay index and travel shortening time, and the calculation process of the delay index of the coordination route is as follows:

the delay index of the road is the ratio of the real-time passing time of the passing route to the passing time of the passing route in a smooth state, and is calculated according to the following formula:

d＝t₁/t₂

wherein d is a delay index of the road; t is t₁The real-time road passing time of the route is the unit of second; t is t₂The time of passing through a route in a smooth state is expressed in seconds; the real-time passing time of the passing route refers to the ratio of the distance of the passing route to the design speed of the route under the condition of free flow, namely, the real-time passing time is calculated according to the following formula:

t₁＝∑s/∑v₁

wherein, t₁The real-time road passing time of the route is the unit of second; s is the distance through the route in km; v. of₁The unit is km/h of the designed speed of a passing route; under the condition of free flow, the design speeds of all road sections of the route are different, and then summation is carried out during subsection calculation of different sections on the route;

the real-time road passing time of the passing route refers to the ratio of the distance of the passing route to the design speed of the route in a smooth state, namely the real-time road passing time is calculated according to the following formula:

t₂＝∑s/∑v₂

wherein, t₂The time is the time of passing through a route in a smooth state, and the unit is second; s is the distance through the route in km; v2 refers to the real-time average speed through the route, with the unit being km/h;

the calculation process of the predicted travel shortening time of the coordination route is as follows:

assuming that the vehicles uniformly arrive at the intersection before coordination, and the vehicles pass through the intersection without stopping after coordination; the predicted travel shortening time of each road section is the difference value between the travel time before coordination and the travel time after coordination;

t_r＝t_{fruit of Chinese wolfberry}-t_{Coordination device}

Wherein, t_rIs the predicted travel shortening time in seconds; t is t_{Fruit of Chinese wolfberry}Is the travel time of the passing route before coordination, the unit is second; t is t_{Coordination device}Is the travel time of the coordinated passing route, and the unit is second; finally, the predicted travel shortening time is displayed, and the predicted travel shortening times of all the road sections need to be added and summed.

7. The method of claim 6, wherein the method comprises: and evaluating the effects of the coordinated route and the coordinated scheme thereof, wherein the effects comprise a recommendation index, the recommendation index of the coordinated route is analyzed according to three indexes, namely a ratio of the predicted travel shortening time to the route length, a ratio of the predicted parking times to the number of coordinated intersections and a delay index, and the recommended ratio analysis result of the coordinated route is processed according to quartile points (0, 0.25,0.5 and 0.75).

8. The method of claim 1, wherein the method comprises: evaluating the effect of the coordination route and the coordination scheme thereof, wherein the evaluation comprises a matching degree which comprises the following steps:

(I) coordinating the matching degree of the recommended phase and the corresponding phase of the intersection with large flow: comparing the coordination phase recommended by the algorithm with the intersection large-flow phase, taking the times of consistent phases as numerators, and taking the total times of comparison as denominators to obtain the matching degree of the recommended coordination phase and the intersection large-flow corresponding phase;

(II) coordinating the matching degree of the recommended phase difference and the road section travel time: for a section of road L_i，jThe matching degree M of the subsection judgment index expression and the coordinated recommended phase difference of the road section and the travel time of the road section_otIs calculated as follows:

wherein M is_otRecommending the matching degree of the phase difference and the road section travel time for coordination; k is a radical of_tRecommending a time period for the coordination;

is a section of road L_i，jThe number of recommended solutions;

is a section of road L_i，jLength of (km);

is k_tSection L of a time segment_i，jUniform velocity (km/h); o is_j(k_t) Recommended section of road L for time period_i，jThe phase difference of (a);

(III) matching degree of the coordination recommendation period and the total flow of the coordination route: for the coordination route j, the subsection judgment index expression of the route and the matching degree M of the coordination recommendation period and the total flow of the coordination route_cvIs calculated as follows:

in the formula, M_cvMatching degree of the coordination recommendation period and the total flow of the coordination route; k is a radical of_tRecommending a time period for the coordination; n is the total number of the coordinated recommendation schemes of the road section; f. of_sum(k_t) Flow sum (vehicle) for the coordinated route in the coordinated recommended time period; c (k)_t) A recommendation period in a coordinated recommendation time period;

(IV) coordinating the matching degree of the recommended green light time and the flow of the coordination scheme: for intersection I_iThe matching degree M of the segmentation judgment index expression and the coordinated recommended green light time of the intersection and the total flow of the coordinated route_gvThe calculation is as follows:

wherein M is_gvRecommending the matching degree of the green light time and the total flow of the coordination route for coordination; k is a radical of_tRecommending a time period for the coordination; n is_iIs a road junction I_iTotal number of recommended solutions;

is a road junction I_iFlow (vehicle) at a coordination phase of a coordination recommendation period;

is a road junction I_iAt k_tGreen time of the coordinated phase of the time segments.

9. A recommendation system for a coordination route and a coordination scheme of urban traffic is characterized by comprising a data module, an algorithm module, a calculation coordination index module, a coordination effect evaluation module and an interaction visualization module;

the data module is used for collecting and preprocessing traffic data;

the algorithm module obtains a recommended coordination route through a threshold judgment method, selects a target road section for the determined coordination route, and establishes a state equation of the change of the flow relation of the target road section along with time; extracting state parameters, control parameters, state time-varying parameters and control time-varying parameters in a state equation of the target road section, constructing an optimal control system of the target road section based on a DDALQR algorithm, and solving optimal time-varying parameters of the target road section, wherein the optimal time-varying parameters describe optimal current-time control parameters under current-time state parameters; calculating the coordinated green light time of the coordinated intersections at the two ends of the target road section at the current moment according to the optimal time-varying parameter of the target road section and the state of the target road section at the current moment; extracting state parameters, control parameters, state time-varying parameters and control time-varying parameters in a state equation of the target road section;

target road section L_i，jEquation of state (1) of (a) translates to:

then the state parameter is x_i，j(k+1)、x_i，j(k)；

The control parameter is

The state time varying parameter is 1;

controlling a time-varying parameter S_i，j+1(k) S_i，j+2(k) S_i，j+3(k) -S_i，j(k)]；

Constructing an optimal control system of a target road section based on a DDALQR algorithm;

converting the state equation of the target road section into a state equation in a DDALQR algorithm:

x(k+1)＝A(k)x(k)+B(k)Δg(k) (3)

wherein x (k +1) and x (k) are the flow of the target road section at the moment k +1 and the moment k; a (k) is a state time-varying parameter at time k; b (k) is a control time-varying parameter at time k;

constructing A (k), and obtaining a time-varying parameter A (k) as an identity matrix of 1 order:

A(k)＝1 (4)

so time varying parameter B_i，j+4(k)：

B_i，j+4(k)＝[S_i，j+1(k) S_i，j+2(k) S_i，j+3(k) -S_i，j+4(k)] (5)

The calculation formula of the control performance index J of the optimal control system is as follows:

wherein the content of the first and second substances,

is the optimal control quantity; q (k) is a traffic state weighting matrix at the moment k; r (k) is a traffic state control weighting matrix at the moment k; the optimal time-varying parameter l (k) that satisfies the performance index J satisfies the following relationship:

wherein the content of the first and second substances,

green time on a phase basis;

obtaining an optimal time-varying parameter of the target road section, wherein the optimal time-varying parameter describes an optimal current-time control parameter under a current-time state parameter;

first by determining the optimum control quantity

The control performance index J in the formula (6) is enabled to be extremely small;

secondly for the section of road L_i，jTo minimize the first term in the control performance index of equation (6), the first term is first transformed into:

wherein each term of the structure Q (x) is

The denominator of Q (x) is minimized for Q (x); thus, the expression for the time-varying parameter q (x) is obtained as:

in the formula: c (k) represents the value for the bearing capacity of the time-varying road section, specifically the historical maximum flow;

for road section L, the same principle applies_i，jIn order to minimize the second term in the control performance index of equation (6), the second term is first used

The transformation is:

wherein each of the structures R (x) is

The molecule of R (x) is minimized, and thereforeObtaining a time-varying parameter R (k):

in summary, the parameters a (k), b (k), q (k), and r (k) are obtained through the equations (4), (5), (10), and (12), and the optimal time-varying parameter l (k) can be calculated by solving the optimization problem with the equation (8) as the objective function and the equations (2), (7), and (6) as the constraint conditions;

the calculation coordination index module is used for calculating a coordination time interval, a coordination phase difference, a coordination cycle and a coordination scheme of a coordination route;

the coordination effect evaluation module is used for evaluating the coordination route and the effect of the coordination scheme thereof to obtain the delay index, the predicted travel shortening time and the coordination recommendation index of the coordination route and the matching degree of the coordination scheme; the interactive visualization module is used for displaying, and comprises a large screen display and each module page; the module displays the whole or regional road network on a large screen, can click to check the details of the intersection, and can dynamically update the speed of the road section, the efficiency index and the predicted travel shortening time; after the coordination route number on the left side is clicked, a coordination time period and a recommended index star level are given, the position of the coordination route is displayed on the map of the main interface on the right side, and the total travel, the speed, the efficiency index and the predicted travel shortening time of the coordination route are displayed under the large screen.

Images