CN111383443A - Method for realizing real-time signal timing control based on dynamic exponential weighted prediction - Google Patents
Method for realizing real-time signal timing control based on dynamic exponential weighted prediction Download PDFInfo
- Publication number
- CN111383443A CN111383443A CN201811611166.1A CN201811611166A CN111383443A CN 111383443 A CN111383443 A CN 111383443A CN 201811611166 A CN201811611166 A CN 201811611166A CN 111383443 A CN111383443 A CN 111383443A
- Authority
- CN
- China
- Prior art keywords
- time
- calculating
- phase
- following
- green
- Prior art date
- Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
- Withdrawn
Links
Images
Classifications
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/01—Detecting movement of traffic to be counted or controlled
- G08G1/0104—Measuring and analyzing of parameters relative to traffic conditions
- G08G1/0125—Traffic data processing
- G08G1/0129—Traffic data processing for creating historical data or processing based on historical data
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/065—Traffic control systems for road vehicles by counting the vehicles in a section of the road or in a parking area, i.e. comparing incoming count with outgoing count
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/07—Controlling traffic signals
-
- G—PHYSICS
- G08—SIGNALLING
- G08G—TRAFFIC CONTROL SYSTEMS
- G08G1/00—Traffic control systems for road vehicles
- G08G1/07—Controlling traffic signals
- G08G1/08—Controlling traffic signals according to detected number or speed of vehicles
Landscapes
- Physics & Mathematics (AREA)
- General Physics & Mathematics (AREA)
- Chemical & Material Sciences (AREA)
- Analytical Chemistry (AREA)
- Traffic Control Systems (AREA)
Abstract
The invention relates to a method for realizing real-time signal timing control based on dynamic exponential weighted prediction, which comprises the steps of (1) judging the current time t1Whether or not greater than tnAnd is not less than t0If yes, acquiring continuous n-stroke periodic flow data; otherwise, exiting the method; (2) predicting next time next flow data F of intersection lanein(ii) a (3) Calculating a flow ratio and Y; (4) judging whether the flow ratio and the Y are less than 0.9, if so, continuing to the step (5); otherwise, continuing the step (6); (5) (6) effective green duration g according to each phaseeiAnd display green time giAnd minimum green time period gminCalculating the total time length t of each phaseiAnd (4) continuing to the step (1). By adopting the method, based on a big Baokang data platform, the traffic flow rule of the intersection can be mastered more accurately through a large amount of historical data; according to history, the exponential weighting algorithm weight is solved in real time through linear programming, and the weight value is dynamically adjusted, so that the calculation result is more accurate, and the personnel investment is reduced.
Description
Technical Field
The invention relates to the field of urban intelligent traffic, in particular to the field of intelligent signal control of urban intelligent traffic, and specifically relates to a method for realizing real-time signal timing control based on dynamic index weighted prediction.
Background
With the gradual acceleration of the urbanization process, the urban traffic problem becomes an urgent problem in China, and more phenomena show that urban traffic congestion is often highlighted at urban road intersections, and the traffic capacity of a plurality of plane intersections is less than 50% of the average traffic capacity of related road sections. Therefore, the key to the full utilization of road resources is the utilization of intersection resources. The traffic signal control system is an extremely important component in the field of modern intelligent traffic. By using an advanced traffic signal control system, the traffic flow can be effectively managed, and the smooth level of urban roads is improved. Various advanced road traffic management schemes are finally realized by means of a traffic signal control system. At present, traffic signal machines are generally used for managing intersections in the traffic management of large and medium cities in China.
At present, a signal control and tuning method is simple, and timing calculation is carried out mainly according to a traffic management department or traffic optimization practitioner through on-site simple data investigation and a traffic engineering method. The reliability of the method mainly depends on the subjective experience of traffic management departments or traffic optimization practitioners, and meanwhile, simple traffic data investigation cannot quantitatively and objectively reflect the traffic flow law of the signal control intersection.
Based on a big data analysis platform, the invention controls a large amount of historical data of the intersection through signals, dynamically adjusts the weight setting of an intersection prediction algorithm (exponential weighting), so that the intersection traffic flow prediction data is more accurate, and meanwhile, an intersection timing scheme is generated through a signal timing method; the invention not only can more accurately play the scientificity of traffic management, but also can reduce the number of employees according to traffic management departments or traffic optimization, and improve the high efficiency and intellectualization of traffic management.
Disclosure of Invention
The invention aims to overcome the defects of the prior art and provides a method for realizing real-time signal timing control based on dynamic index weighted prediction, which has the advantages of prediction accuracy, high efficiency and intellectualization.
In order to achieve the above object, the method for realizing real-time signal timing control based on dynamic exponential weighted prediction of the present invention comprises the following steps:
the method for realizing real-time signal timing control based on the system and based on dynamic index weighted prediction is mainly characterized by comprising the following steps of:
(1) obtaining an operating time [ t ]0，tn]Intersection lane data LiCurrent time t1Judging the current time t1Whether or not greater than tnAnd is not less than t0If yes, acquiring continuous n-stroke periodic flow data, and converting into hourly flow; otherwise, exiting the method;
(2) calculating an exponential weighting dynamic parameter value β, predicting the next time flow data F of the intersection lane according to the exponential weightingin；
(3) Obtaining the current signal stage parameters of the intersection and calculating the traffic capacity C of the laneiAnd calculating the maximum flow ratio y of the phase lanejAnd the flow ratio and Y;
(4) judging whether the flow ratio and the Y are less than 0.9, if so, continuing to the step (5); otherwise, continuing the step (6);
(5) according to the period green light loss time L and the period duration C0Calculating the effective green light duration G of the periodEAnd according to the effective green duration g of each phaseeiAnd display green time giAnd minimum green time period gminCalculating the total time length t of each phaseiContinuing the step (1);
(6) calculating the ratio of each phase flow ratio to the sum of the flow ratios according to the period green light loss time L and the period
And according to the effective green duration g of each phaseeiAnd display green time giAnd minimum green time period gminCalculating the total time length t of each phaseiContinuing step (1)
Preferably, the reduced hourly flow rate F in the step (1)iThe method specifically comprises the following steps:
the hourly flow F is reduced according to the following formulai：
Fi＝3600*fi/Ti；
Wherein f isiThe flow data is n continuous periods of flow data, and T is the period length.
Preferably, the step (2) specifically comprises the following steps:
(2.1) calculating an exponentially weighted prediction result Pn；
(2.2) constructing the loss function LminObtaining the optimal hyper-parameter theta by a branch definition method0、θ1、θ2、...、θkAnd an β value was obtained;
(2.3) calculating the next period flow F of all lanes at the intersection according to the exponential weightingin。
Preferably, the exponentially weighted prediction result P in step (2.1) is calculatednThe method specifically comprises the following steps:
calculating an exponentially weighted prediction result P according to the following formulan：
Pn＝θ0*Fn-1+θ1*Fn-2+θ2*Fn-3+...+θk*Fn-(k+1)；
Wherein, theta0、θ1、θ2、...、θkFor exponentially weighting the hyperparameters, Fn-1、Fn-2、Fn-3、...、Fn-(k+1)Is the actual value.
Preferably, the loss function L is constructed in the step (2.2)minThe method specifically comprises the following steps:
constructing the loss function L according to the following formulamin：
And satisfies the following conditions:
wherein, PnFor exponentially weighting the prediction results, FnIs the actual value.
Preferably, the step (2.3) of calculating the next cycle flow F of all lanes at the intersectioninThe method specifically comprises the following steps:
calculating the next period flow F of all lanes of the intersection according to the following formulain：
Fin＝F0*(1-β)+F1*β*(1-β)+F2*β2*(1-β)+F3*β3(1-β)+.....+Fn-1*βN-1*(1-β)；
Wherein β is an exponentially weighted dynamic parameter value.
Preferably, the phase lane maximum flow ratio y calculated in the step (3) isjThe method specifically comprises the following steps:
calculating the phase lane maximum flow ratio y according to the following formulaj：
yj＝max[Fin/Ci]；
Wherein i is the lane number and j is the phase number.
Preferably, the calculating the flow ratio and Y in the step (3) specifically includes:
the flow ratio and Y are calculated according to the following formulas:
Y＝∑yj；
wherein j is the phase number.
Preferably, the calculating of the period green light loss time L in the step (5) includes:
the cycle green light loss time L is calculated according to the following formula:
L＝∑K(LS+I-A)；
wherein L issFor signal loss per phaseLosing time, I is the green light interval time, and A is the yellow light time.
Preferably, the period duration C is calculated in the step (5)0The method specifically comprises the following steps:
the cycle duration C is calculated according to the following formula0：
Preferably, said calculating the period valid green light duration G in step (5)EThe method specifically comprises the following steps:
calculating the period valid green time duration G according to the following formulaE：
GE＝C0-L。
Preferably, said step (5) of calculating the effective green duration g of each phaseeiThe method specifically comprises the following steps:
calculating the effective green time length g of each phase according to the following formulaei：
gei＝GE*yi；
Wherein, yiThe maximum lane flow ratio for phase i.
Preferably, the calculation in step (5) displays the green time giThe method specifically comprises the following steps:
calculating and displaying green light time g according to the following formulai：
gi＝gei+II-Ai；
Wherein, IIGreen interval time for phase i, AiIs the start-up loss time for phase i.
Preferably, the shortest green light time length g is calculated in the step (5)minThe method specifically comprises the following steps:
the shortest green time g is calculated according to the following formulaiin：
Wherein L ispLength of pedestrian crossing, VpSpeed of pedestrian crossing.
Preferably, the step (5) calculates the total duration t of each phaseiThe method specifically comprises the following steps:
the total duration t of each phase is calculated according to the following formulai：
ti＝gi+II+Ai。
Preferably, the calculating of the period green light loss time L in the step (6) includes:
the cycle green light loss time L is calculated according to the following formula:
L＝∑K(LS+I-A)；
wherein L issFor signal loss time per phase, I is the green light interval time and a is the yellow light time.
Preferably, the period duration C is calculated in the step (6)0The method specifically comprises the following steps:
the cycle duration C is calculated according to the following formula0：
C0＝T0；
Wherein, T0Is the original scheme period.
Preferably, the step (6) calculates a ratio of each phase flow ratio to the sum of the flow ratios
The method specifically comprises the following steps: calculating the ratio of each phase flow ratio to the sum of the flow ratios according to the following formula
Preferably, said step (6) of calculating the effective green duration g of each phaseeiThe method specifically comprises the following steps:
calculating the effective green time length g of each phase according to the following formulaei：
Wherein, T0Is the original scheme period.
Preferably, the calculation in step (6) displays the green time giThe method specifically comprises the following steps:
calculating and displaying green light time g according to the following formulai：
gi＝gei+II-Ai；
Preferably, the shortest green light time length g is calculated in the step (6)minThe method specifically comprises the following steps:
the shortest green time g is calculated according to the following formulamin：
Wherein L ispLength of pedestrian crossing, VpThe speed of pedestrian crossing the street, I is the green light interval time.
Preferably, the step (6) of calculating the total duration t of each phaseiThe method specifically comprises the following steps:
the total duration t of each phase is calculated according to the following formulai：
ti＝gi+II+Ai。
By adopting the method for realizing real-time signal timing control based on dynamic index weighted prediction, the intersection traffic flow rule can be more accurately mastered through a large amount of historical data based on a big Baokang data platform; according to history, solving the weight of the exponential weighting algorithm in real time through linear programming, and dynamically adjusting the weight value to enable the calculation result to be more accurate; generating a next period signal timing scheme by combining signal control intersection real-time prediction data through a signal timing method; the highly intelligent control system reduces personnel investment.
Drawings
Fig. 1 is an overall flowchart of a method for implementing real-time signal timing control based on dynamic exponential weighted prediction according to the present invention.
Detailed Description
In order to more clearly describe the technical contents of the present invention, the following further description is given in conjunction with specific embodiments.
The invention discloses a method for realizing real-time signal timing control based on dynamic exponential weighted prediction, which comprises the following steps:
(1) obtaining an operating time [ t ]0，tn]Intersection lane data LiCurrent time t1Judging the current time t1Whether or not greater than tnAnd is not less than t0If yes, acquiring continuous n-stroke periodic flow data, and converting into hourly flow; otherwise, exiting the method;
(2) calculating an exponential weighting dynamic parameter value β, predicting the next time flow data F of the intersection lane according to the exponential weightingin；
(2.1) calculating an exponentially weighted prediction result Pn；
(2.2) constructing the loss function LminObtaining the optimal hyper-parameter theta by a branch definition method0、θ1、θ2、...、θkAnd an β value was obtained;
(2.3) calculating the next period flow F of all lanes at the intersection according to the exponential weightingin；
(3) Obtaining the current signal stage parameters of the intersection and calculating the traffic capacity C of the laneiAnd calculating the maximum flow ratio y of the phase lanejAnd the flow ratio and Y;
(4) judging whether the flow ratio and the Y are less than 0.9, if so, continuing to the step (5); otherwise, continuing the step (6);
(5) according to the period green light loss time L and the period duration C0Calculating the effective green light duration G of the periodEAnd according to the effective green duration g of each phaseeiAnd display green time giAnd minimum green time period gminCalculating the total time length t of each phaseiContinuing the step (1);
(6) green light loss time according to period L andperiodically calculating the ratio of each phase flow ratio to the sum of the flow ratios
And according to the effective green duration g of each phaseeiAnd display green time giAnd minimum green time period gminCalculating the total time length t of each phaseiContinuing step (1)
As a preferred embodiment of the present invention, the reduced hourly flow rate F in the step (1) isiThe method specifically comprises the following steps:
the hourly flow F is reduced according to the following formulai：
Fi＝3600*fi/Ti；
Wherein f isiThe flow data is n continuous periods of flow data, and T is the period length.
As a preferred embodiment of the present invention, the calculation of the exponentially weighted prediction result P in the step (2.1) is performednThe method specifically comprises the following steps:
calculating an exponentially weighted prediction result P according to the following formulan：
Pn＝θ0*Fn-1+θ1*Fn-2+θ2*Fn-3+...+θk*Fn-(k+1)；
Wherein, theta0、θ1、θ2、...、θkFor exponentially weighting the hyperparameters, Fn-1、Fn-2、Fn-3、...、Fn-(k+1)Is the actual value.
As a preferred embodiment of the present invention, the figure loss function L in the step (2.2) isminThe method specifically comprises the following steps:
constructing the loss function L according to the following formulamin：
And satisfies the following conditions:
wherein, PnFor exponentially weighting the prediction results, FnIs the actual value.
As a preferred embodiment of the present invention, the step (2.3) of calculating the next cycle flow F of all lanes at the intersectioninThe method specifically comprises the following steps:
calculating the next period flow F of all lanes of the intersection according to the following formulain：
Fin＝F0*(1-β)+F1*β*(1-β)+F2*β2*(1-β)+F3*β3(1-β)+.....+Fn-1*βN-1*(1-β)；
Wherein β is an exponentially weighted dynamic parameter value.
As a preferred embodiment of the present invention, the phase lane maximum flow rate ratio y calculated in the step (3) isjThe method specifically comprises the following steps:
calculating the phase lane maximum flow ratio y according to the following formulaj：
yj＝max[Fin/Ci]；
Wherein i is the lane number and j is the phase number.
As a preferred embodiment of the present invention, the calculating the flow ratio and Y in the step (3) specifically includes:
the flow ratio and Y are calculated according to the following formulas:
Y＝∑yj；
wherein j is the phase number.
As a preferred embodiment of the present invention, the calculating period green light loss time L in step (5) specifically includes:
the cycle green light loss time L is calculated according to the following formula:
L＝∑K(LS+I-A)；
wherein L issFor signal loss time per phase, I is the green light interval time and a is the yellow light time.
As a preferred embodiment of the present inventionThe period duration C is calculated in the step (5)0The method specifically comprises the following steps:
the cycle duration C is calculated according to the following formula0：
In a preferred embodiment of the present invention, the period valid green light duration G is calculated in step (5)EThe method specifically comprises the following steps:
calculating the period valid green time duration G according to the following formulaE：
GE＝C0-L。
In a preferred embodiment of the present invention, the step (5) is to calculate the effective green duration g of each phaseeiThe method specifically comprises the following steps:
calculating the effective green time length g of each phase according to the following formulaei：
gei＝GE*yi；
Wherein, yiThe maximum lane flow ratio for phase i.
As a preferred embodiment of the present invention, the calculation in the step (5) displays the green time giThe method specifically comprises the following steps:
calculating and displaying green light time g according to the following formulai：
gi＝gei+II-Ai；
Wherein, IIGreen interval time for phase i, AiIs the start-up loss time for phase i.
As a preferred embodiment of the present invention, the shortest green time period g is calculated in the step (5)minThe method specifically comprises the following steps:
the shortest green time g is calculated according to the following formulamin：
Wherein L ispLength of pedestrian crossing, VpThe speed of pedestrian crossing the street, I is the green light interval time.
As a preferred embodiment of the present invention, the step (5) of calculating the total duration t of each phaseiThe method specifically comprises the following steps:
the total duration t of each phase is calculated according to the following formulai：
ti＝gi+II+Ai。
As a preferred embodiment of the present invention, the calculating period green light loss time L in step (6) specifically includes:
the cycle green light loss time L is calculated according to the following formula:
L＝∑K(LS+I-A)；
wherein L issFor signal loss time per phase, I is the green light interval time and a is the yellow light time.
As a preferred embodiment of the present invention, the period duration C is calculated in the step (6)0The method specifically comprises the following steps:
the cycle duration C is calculated according to the following formula0：
C0＝T0；
Wherein, T0Is the original scheme period.
In a preferred embodiment of the present invention, the step (6) calculates a ratio of each phase flow rate ratio to the sum of the flow rate ratios
The method specifically comprises the following steps:
calculating the ratio of each phase flow ratio to the sum of the flow ratios according to the following formula
As a preferred embodiment of the present invention, in the step (6), theCalculating the effective green light duration g of each phaseeiThe method specifically comprises the following steps:
calculating the effective green time length g of each phase according to the following formulaei：
Wherein, T0Is the original scheme period.
As a preferred embodiment of the present invention, the calculation in the step (6) displays the green time giThe method specifically comprises the following steps:
calculating and displaying green light time g according to the following formulai：
gi＝gei+II-Ai；
As a preferred embodiment of the present invention, the shortest green time period g is calculated in the above-mentioned step (6)minThe method specifically comprises the following steps:
the shortest green time g is calculated according to the following formulamin：
Wherein L ispLength of pedestrian crossing, VpSpeed of pedestrian crossing.
As a preferred embodiment of the present invention, the step (6) of calculating the total duration t of each phaseiThe method specifically comprises the following steps:
the total duration t of each phase is calculated according to the following formulai：
ti＝gi+II+Ai。
In the specific implementation mode of the invention, based on a big data analysis platform, a large amount of historical data of the intersection is controlled by a signal, and the weight setting of an intersection prediction algorithm (exponential weighting) is dynamically adjusted, so that the intersection traffic flow prediction data is more accurate, and an intersection timing scheme is generated by a signal timing method; the invention not only can more accurately play the scientificity of traffic management, but also can reduce the number of employees according to traffic management departments or traffic optimization, and improve the high efficiency and intellectualization of traffic management.
The invention predicts the intersection traffic flow by an index weighting method of dynamic weight based on extracting the historical intersection flow data; and calculating a signal timing scheme in real time according to the obtained flow value of each lane of the intersection. The specific operation logic is as follows:
1. obtaining an operating time [ t ]0、tn]
2. Extracting intersection lane data Li；
3. Obtaining the current time t1；
4. If t1∈[t0、tn]Starting the algorithm, otherwise exiting;
5. extracting intersection lane LiN continuous periodic flow data f before current time0、f1、、、fn-1；
6. Will f isiReduced hourly flow Fi＝3600*fi/Ti；
7. The exponentially weighted dynamic parameter values β are calculated as follows:
Pn＝θ0*Fn-1+θ1*Fn-2+θ2*Fn-3+...+θk*Fn-(k+1)
Pnexponentially weighting the prediction results;
θ0、θ1、θ2、...、θkexponentially weighting the hyper-parameters;
Fn-1、Fn-2、Fn-3、...、Fn-(k+1)is an actual value;
to find the optimal set of hyper-parameters, a loss function is constructed:
and satisfies the following conditions
Solving the planning problem with conditions by using a branch definition method to obtain the optimal hyper-parameter theta0、θ1、θ2、...、θkFurther, β values may be determined.
8. Then predicting the next time L of the lane at the intersection by exponential weightingiNext flow data Fin：
Fn＝F0*(1-β)+F1*β*(1-β)+F2*β2*(1-β)+F3*β3(1-β)+.....+Fn-1*βN-1*(1-β)；
9. Calculating the next period flow F of all lanes of the intersection according to the step (8)in；
10. Acquiring parameters of a current signal stage of the intersection;
11. the traffic capacity C of the lanei；
12. Calculating the maximum flow ratio y of the phase lanej＝max[Fin/Ci](ii) a (i represents a lane number, j represents a phase number)
13. Calculating the flow ratio and Y- ∑ Yj；
14. If Y is less than 0.9, continuing the steps (15) to (22), otherwise, entering the steps (23) to (31);
15. calculating the green light loss time L- ∑K(LS+ I-A); (signal loss time per phase is LsGreen light interval time I, yellow light time A)
16. Calculating cycle duration
17. Calculating the effective green light duration G of the periodE＝C0-L；
18. Calculating the effective green light duration g of each phaseei＝Ge*yi
19. Calculating the shortest duration of green light
(pedestrian crossing length LPThe speed of pedestrian crossing is VP)
20. Calculating and displaying green light time gi＝gei+II-Ai；
21. When g isi≥gmin，gi＝giWhen g isi＜gmin，gi＝gmin；
22. Output the result ti＝gi+II+Ai；(tiFor the total duration of each phase, including yellow flash and full red)
23. If Y is more than or equal to 0.9, continuing the next step;
24. calculating the green light loss time L- ∑K(LS+ I-A); (signal loss time per phase is LsGreen light interval time I, yellow light time A)
25. Setting period C0＝T0；(T0As the original recipe period);
26. calculating the ratio of each phase flow ratio to the sum of the flow ratios
27. Calculating the effective green time duration of the given phase
28. Calculating the shortest duration of green light
(pedestrian crossing length LPThe speed of pedestrian crossing is VP)
29. Calculating and displaying green light time gi＝gei+II-Ai；
30. When g isi≥gmin，gi＝giWhen g isi＜gmin，gi＝gmin；
31. Output the result ti＝gi+II+Ai；(tiFor the total duration of each phase, including yellow flash and full red)
32. Entering the next calculation period to obtain the current time ti；
33. If ti∈[t0、tn]And (5) continuously repeating the steps (5) - (21), and otherwise, exiting.
By adopting the method for realizing real-time signal timing control based on dynamic index weighted prediction, the intersection traffic flow rule can be more accurately mastered through a large amount of historical data based on a big Baokang data platform; according to history, solving the weight of the exponential weighting algorithm in real time through linear programming, and dynamically adjusting the weight value to enable the calculation result to be more accurate; generating a next period signal timing scheme by combining signal control intersection real-time prediction data through a signal timing method; the highly intelligent control system reduces personnel investment.
In this specification, the invention has been described with reference to specific embodiments thereof. It will, however, be evident that various modifications and changes may be made thereto without departing from the broader spirit and scope of the invention. The specification and drawings are, accordingly, to be regarded in an illustrative rather than a restrictive sense.
Claims (22)
1. A method for realizing real-time signal timing control based on dynamic exponential weighted prediction is characterized by comprising the following steps:
(1) obtaining an operating time [ t ]0，tn]Intersection lane data LiCurrent time t1Judging the current time t1Whether or not greater than tnAnd is not less than t0If yes, acquiring continuous n-stroke periodic flow data, and converting into hourly flow; otherwise, exiting the method;
(2) calculating an exponential weighting dynamic parameter value β, predicting the next time flow data F of the intersection lane according to the exponential weightingin；
(3) Obtaining the current signal stage parameters of the intersection and calculating the traffic capacity C of the laneiAnd calculating the maximum flow ratio y of the phase lanejAnd the flow ratio and Y;
(4) judging whether the flow ratio and the Y are less than 0.9, if so, continuing to the step (5); otherwise, continuing the step (6);
(5) according to the period green light loss time L and the period duration C0Calculating the effective green light duration G of the periodEAnd according to the effective green duration g of each phaseeiAnd display green time giAnd minimum green time period gminCalculating the total time length t of each phaseiContinuing the step (1);
(6) calculating the ratio theta of each phase flow ratio to the sum of the flow ratios according to the period green light loss time L and the periodiAnd according to the effective green duration g of each phaseeiAnd display green time giAnd minimum green time period gminCalculating the total time length t of each phaseiAnd (4) continuing to the step (1).
2. The method for implementing real-time signal timing control based on dynamic exponential weighted prediction as claimed in claim 1, wherein the reduced hourly flow rate F in step (1)iThe method specifically comprises the following steps:
the hourly flow F is reduced according to the following formulai：
Fi＝3600*fi/Ti；
Wherein f isiThe flow data is n continuous periods of flow data, and T is the period length.
3. The method according to claim 1, wherein the step (2) comprises the following steps:
(2.1) calculating an exponentially weighted prediction result Pn；
(2.2) constructing the loss function LminObtaining the optimal hyper-parameter theta by a branch definition method0、θ1、θ2、...、θkAnd an β value was obtained;
(2.3) calculating the next period flow F of all lanes at the intersection according to the exponential weightingin。
4. The method of claim 3, wherein said step (2.1) of calculating the exponentially weighted prediction result PnThe method specifically comprises the following steps:
calculating an exponentially weighted prediction result P according to the following formulan：
Pn＝θ0*Fn-1+θ1*Fn-2+θ2*Fn-3+...+θk*Fn-(k+1)；
Wherein, theta0、θ1、θ2、...、θkFor exponentially weighting the hyperparameters, Fn-1、Fn-2、Fn-3、...、Fn-(k+1)Is the actual value.
5. The method of claim 3 wherein said step (2.2) is performed by constructing a loss function LminThe method specifically comprises the following steps:
constructing the loss function L according to the following formulamin：
And satisfies the following conditions:
wherein, PnFor exponentially weighting the prediction results, FnIs the actual value.
6. The method for realizing real-time signal timing control based on dynamic exponential weighted prediction as claimed in claim 3, wherein the step (2.3) of calculating the next periodic flow F of all lanes of the intersectioninThe method specifically comprises the following steps:
calculating the next of all lanes of the intersection according to the following formulaPeriodic flow Fin：
Fin＝F0*(1-β)+F1*β*(1-β)+F2*β2*(1-β)+F3*β3(1-β)+.....+Fn-1*βN-1*(1-β)；
Wherein β is an exponentially weighted dynamic parameter value.
7. The method for implementing real-time signal timing control based on dynamic exponential weighted prediction as claimed in claim 1, wherein the step (3) of calculating the maximum flow ratio y of the phase lanejThe method specifically comprises the following steps:
calculating the phase lane maximum flow ratio y according to the following formulaj：
yj＝max[Fin/Gi]；
Wherein i is the lane number and j is the phase number.
8. The method for realizing real-time signal timing control based on dynamic exponential weighted prediction as claimed in claim 1, wherein the calculating the flow ratio and Y in the step (3) specifically comprises:
the flow ratio and Y are calculated according to the following formulas:
Y＝∑yj；
wherein j is the phase number.
9. The method for implementing real-time signal timing control based on dynamic exponential weighted prediction as claimed in claim 1, wherein the calculating of the period green light loss time L in the step (5) specifically comprises:
the cycle green light loss time L is calculated according to the following formula:
L＝∑K(Ls+I-A)；
wherein L issFor signal loss time per phase, I is the green light interval time and a is the yellow light time.
10. The dynamic index addition based on claim 1The method for realizing real-time signal timing control by weight prediction is characterized in that the period duration C is calculated in the step (5)0The method specifically comprises the following steps:
the cycle duration C is calculated according to the following formula0：
11. The method for implementing real-time signal timing control based on dynamic exponentially weighted prediction as claimed in claim 1, wherein said step (5) is performed to calculate the period valid green time duration GEThe method specifically comprises the following steps:
calculating the period valid green time duration G according to the following formulaE：
GE＝C0-L。
12. The method for implementing real-time signal timing control based on dynamic exponential weighted prediction as claimed in claim 1, wherein said step (5) of calculating the effective green duration g of each phaseeiThe method specifically comprises the following steps:
calculating the effective green time length g of each phase according to the following formulaei：
gei＝GE*yi；
Wherein, yiThe maximum lane flow ratio for phase i.
13. The method for implementing real-time signal timing control based on dynamic exponentially weighted prediction as claimed in claim 1, wherein said calculating in step (5) displays a green time giThe method specifically comprises the following steps:
calculating and displaying green light time g according to the following formulai：
gi＝gei+I-Ai；
Wherein, IIGreen interval time for phase i, AiIs the start-up loss time for phase i.
14. The method for implementing real-time signal timing control based on dynamic exponentially weighted prediction as claimed in claim 1, wherein said step (5) is performed to calculate the shortest green time period gminThe method specifically comprises the following steps:
the shortest green time g is calculated according to the following formulamin：
Wherein L ispLength of pedestrian crossing, VpSpeed of pedestrian crossing.
15. The method for implementing real-time signal timing control based on dynamic exponentially weighted prediction as claimed in claim 1, wherein said step (5) is performed by calculating a total duration t of each phaseiThe method specifically comprises the following steps:
the total duration t of each phase is calculated according to the following formulai：
ti＝gi+II+Ai。
16. The method for realizing real-time signal timing control based on dynamic exponential weighted prediction as claimed in claim 1, wherein the calculating of the period green light loss time L in the step (6) specifically comprises:
the cycle green light loss time L is calculated according to the following formula:
L＝∑K(Ls+I-A)；
wherein L issFor signal loss time per phase, I is the green light interval time and a is the yellow light time.
17. The method for implementing real-time signal timing control based on dynamic exponentially weighted prediction as claimed in claim 1, wherein said step (6) calculates the cycle duration C0The method specifically comprises the following steps:
the cycle duration C is calculated according to the following formula0：
C0＝T0；
Wherein, T0Is the original scheme period.
18. The method for implementing real-time signal timing control based on dynamic exponentially weighted prediction as claimed in claim 1, wherein said step (6) is performed by calculating a ratio of each phase flow ratio to a sum of flow ratios
The method specifically comprises the following steps:
calculating the ratio of each phase flow ratio to the sum of the flow ratios according to the following formula
19. The method for implementing real-time signal timing control based on dynamic exponential weighted prediction as claimed in claim 1, wherein said step (6) of calculating the effective green duration g for each phaseeiThe method specifically comprises the following steps:
calculating the effective green time length g of each phase according to the following formulaei：
Wherein, T0Is the original scheme period.
20. The method for implementing real-time signal timing control based on dynamic exponentially weighted prediction as claimed in claim 1, wherein said calculating in step (6) displays a green time giThe method specifically comprises the following steps:
calculating and displaying green light time g according to the following formulai：
gi＝gei+II-Ai；
21. The method for implementing real-time signal timing control based on dynamic exponentially weighted prediction as claimed in claim 1, wherein said step (6) is performed to calculate the shortest green time period gminThe method specifically comprises the following steps:
the shortest green time g is calculated according to the following formulamin：
Wherein L ispLength of pedestrian crossing, VpThe speed of pedestrian crossing the street, I is the green light interval time.
22. The method for implementing real-time signal timing control based on dynamic exponentially weighted prediction as claimed in claim 1, wherein said step (6) is performed by calculating a total duration t of each phaseiThe method specifically comprises the following steps:
the total duration t of each phase is calculated according to the following formulai：
ti＝gi+II+Ai。
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811611166.1A CN111383443A (en) | 2018-12-27 | 2018-12-27 | Method for realizing real-time signal timing control based on dynamic exponential weighted prediction |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201811611166.1A CN111383443A (en) | 2018-12-27 | 2018-12-27 | Method for realizing real-time signal timing control based on dynamic exponential weighted prediction |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN111383443A true CN111383443A (en) | 2020-07-07 |
Family
ID=71217845
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201811611166.1A Withdrawn CN111383443A (en) | 2018-12-27 | 2018-12-27 | Method for realizing real-time signal timing control based on dynamic exponential weighted prediction |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN111383443A (en) |
Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104036646A (en) * | 2014-06-26 | 2014-09-10 | 公安部交通管理科学研究所 | Method for dividing signal-timing periods of intersections |
| CN109035780A (en) * | 2018-09-07 | 2018-12-18 | 江苏智通交通科技有限公司 | The Optimal Configuration Method in section is incuded when signal controlled junctions phase is green |
-
2018
- 2018-12-27 CN CN201811611166.1A patent/CN111383443A/en not_active Withdrawn
Patent Citations (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN104036646A (en) * | 2014-06-26 | 2014-09-10 | 公安部交通管理科学研究所 | Method for dividing signal-timing periods of intersections |
| CN109035780A (en) * | 2018-09-07 | 2018-12-18 | 江苏智通交通科技有限公司 | The Optimal Configuration Method in section is incuded when signal controlled junctions phase is green |
Non-Patent Citations (2)
| Title |
|---|
| 诸云: "基于交通流模式的交叉口动态信号控制研究", 《中国优秀硕士学位论文全文数据库（工程科技Ⅱ辑）》 * |
| 高佳宁，等: "基于改进F-B模型的过饱和状态下交叉口信号配时优化方法", 《山东科学》 * |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| US10902719B2 (en) | Method of predicting traffic congestion and controlling traffic signals based on deep learning and server for performing the same | |
| CN105788302B (en) | A kind of city traffic signal lamp dynamic timing method of biobjective scheduling | |
| CN103996289B (en) | A kind of flow-speeds match model and Travel Time Estimation Method and system | |
| WO2019047905A1 (en) | Road traffic analysis system, method and apparatus | |
| CN109410574B (en) | Phase-phase signal control scheme-oriented timing parameter optimization method | |
| CN111710167A (en) | Single-point optimization control method and device based on online traffic simulation | |
| CN105303856B (en) | A kind of variable information board information issuing method of prediction model | |
| CN108877253B (en) | Internet of things-based bus lane resource dynamic sharing method and system | |
| CN105761515B (en) | A kind of intersection signal dynamic adjusting method and device, system | |
| CN105046990B (en) | Pedestrian walkway signals' control method between a kind of adjacent intersection based on particle cluster algorithm | |
| CN106017496A (en) | Real-time navigation method based on road condition | |
| CN108629970B (en) | Intersection signal parameter optimization method based on Monte Carlo tree search | |
| CN111524345B (en) | Induction control method for multi-objective optimization under constraint of real-time queuing length of vehicle | |
| CN115171408B (en) | Traffic signal optimization control method | |
| Chen et al. | Dynamic traffic light optimization and Control System using model-predictive control method | |
| CN117351734A (en) | Intelligent regulation and control method and system for vehicle delay | |
| CN118015857B (en) | Road traffic planning method | |
| Fu et al. | Modeling and simulation of dynamic lane reversal using a cell transmission model | |
| Liu | A short-term traffic flow prediction method based on svr | |
| CN107680393B (en) | Intelligent control method of crossroad traffic signal lamp based on time-varying domain | |
| CN111383443A (en) | Method for realizing real-time signal timing control based on dynamic exponential weighted prediction | |
| CN116665444A (en) | Expressway ramp control and variable speed limit cooperative control method considering accident risk | |
| CN115472023B (en) | Intelligent traffic light control method and device based on deep reinforcement learning | |
| CN114863687B (en) | Road traffic intelligent optimization method based on trunk line green wave | |
| CN111383469A (en) | Method for realizing real-time signal calculation function based on big data linear regression prediction algorithm |
Legal Events
| Date | Code | Title | Description |
|---|---|---|---|
| PB01 | Publication | ||
| PB01 | Publication | ||
| SE01 | Entry into force of request for substantive examination | ||
| SE01 | Entry into force of request for substantive examination | ||
| WW01 | Invention patent application withdrawn after publication |
Application publication date: 20200707 |
|
| WW01 | Invention patent application withdrawn after publication |