CN108847021B - Road network flow prediction method considering heterogeneous users - Google Patents

Road network flow prediction method considering heterogeneous users Download PDF

Info

Publication number
CN108847021B
CN108847021B CN201810579045.7A CN201810579045A CN108847021B CN 108847021 B CN108847021 B CN 108847021B CN 201810579045 A CN201810579045 A CN 201810579045A CN 108847021 B CN108847021 B CN 108847021B
Authority
CN
China
Prior art keywords
path
time
flow
day
probability
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.)
Expired - Fee Related
Application number
CN201810579045.7A
Other languages
Chinese (zh)
Other versions
CN108847021A (en
Inventor
田晟
许凯
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
South China University of Technology SCUT
Original Assignee
South China University of Technology SCUT
Priority date (The priority date 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 date listed.)
Filing date
Publication date
Application filed by South China University of Technology SCUT filed Critical South China University of Technology SCUT
Priority to CN201810579045.7A priority Critical patent/CN108847021B/en
Publication of CN108847021A publication Critical patent/CN108847021A/en
Application granted granted Critical
Publication of CN108847021B publication Critical patent/CN108847021B/en
Expired - Fee Related legal-status Critical Current
Anticipated expiration legal-status Critical

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
    • G08G1/0137—Measuring and analyzing of parameters relative to traffic conditions for specific applications
    • 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
    • G08G1/065—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

Landscapes

  • Physics & Mathematics (AREA)
  • General Physics & Mathematics (AREA)
  • Chemical & Material Sciences (AREA)
  • Analytical Chemistry (AREA)
  • Traffic Control Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a road network flow prediction method considering heterogeneous users, which comprises the steps of S1 updating road network parameters according to empirical learning, S2 determining a value function in a foreground theory, and S3 determining a probability weight function; s4, calculating a path foreground value; s5, a departure period flow evolution model with the maximum quasi-point arrival probability as a target; s6, obtaining a path flow evolution model with the path foreground value as the maximum target. The invention can predict the flow of the congested toll road network more accurately.

Description

Road network flow prediction method considering heterogeneous users
Technical Field
The invention relates to the field of traffic, in particular to a road network flow prediction method considering heterogeneous users.
Background
In order to relieve the contradiction between supply and demand of urban traffic, road congestion charging becomes an effective traffic control means by balancing traffic volume in space and time. The existing research mainly focuses on the formulation of an optimal charging strategy, for example, the influence of travel time reliability in a random network system on the path selection behavior is researched, and the higher the reliability required by a traveler is, the less obvious the charging effect is; different charging strategies have been investigated, including staged charging and dynamic charging, and approaching dynamic charging with single-stage charging. The traffic distribution of the path traffic is carried out by using a (random) traveler balance or (random) system optimization, the result is represented as the distribution of the traffic in the day, and the unbalanced process of adjusting the travel decision by the traveler according to the trip experience in the travel process is not considered, so that the traffic is represented dynamically at the departure time and on the travel path.
At present, the situation of road network charging is less considered in the aspect of a day-by-day flow distribution theory, and heterogeneous travelers in a charging road network can generate different selection behaviors for a charging time interval, a charging road section and a charging value, so that different flow evolution phenomena occur. And road tolling is different from an emergency and has repeatability. Therefore, the research on the road network flow evolution of the road congestion has important significance for traffic management and control and the establishment of a charging policy.
Disclosure of Invention
In order to overcome the defects and shortcomings of the prior art, the invention provides a road network traffic prediction method considering heterogeneous users.
The method provides that travelers are classified according to time values, and road network parameters are updated according to an empirical learning mechanism. And establishing a flow transfer model at the departure time by taking the maximum standard point arrival probability as a target, and establishing a path flow transfer model by taking the maximum foreground value as a target.
The invention adopts the following technical scheme:
a road network flow prediction method considering heterogeneous users comprises the following steps:
s1, updating road network parameters according to empirical learning;
s2, determining a cost function in the foreground theory;
s3, determining a probability weight function;
s4, calculating a path foreground value;
s5, a departure period flow evolution model with the maximum quasi-point arrival probability as a target;
s6, obtaining a path flow evolution model with the path foreground value as the maximum target.
The updating of the road network parameters according to the empirical learning specifically comprises the following steps:
the path running time follows normal distribution and has Tk～N(τk,σk),τk，σkRespectively representing the average value and the variance of the walking time of the path k, and updating the parameters of the path k in the m period on the nth day by the data of the same path on the (n-1) th day at the same time period:
τk(n,m)＝(τk(n-1,m)·(n-1)+tk(n,m))/n
σk(n,m)2＝(σk(n-1,m)2·(n-1)+(tk(n,m)-τk(n,m))2)/n
Tkfor the running time distribution function, N denotes the normal distribution, tk(n, m) represents the actual travel time of the kth route in the period of m on the nth day, and when n is 1, tauk(1,m)＝tk free,σk(1,m)＝0,tk freeRepresenting the free running time of the path;
the travelers walking on the path before going out every dayThe travel time has an estimated value defined as the understood travel time etk(n, m), updating the same path of the n-th day time period by the understanding travel time of the m-th day period path k and the actual travel time according to historical travel experience by a traveler:
Figure BDA0001687947200000021
Figure BDA0001687947200000022
is a parameter which reflects the degree of dependence of a traveler on the actual traveling time of the previous day,
Figure BDA0001687947200000023
the smaller, the greater the degree of dependence,
Figure BDA0001687947200000024
the larger, the smaller the degree of dependence,
Figure BDA0001687947200000025
the value range is [0,1 ]]When n is 1, etk(1,m)＝tk free。
The determination of the cost function in the foreground theory in S2 specifically includes:
defining a traffic network G ═ W, A, W is a node set, A is a road segment set, a road segment belongs to A, each day of research time interval is defined as [ T1, T2], each day of charging time interval is defined as [ T1', T2' ], wherein T1 is not less than T1 '< T2' < T2, the research time interval is divided into M equal time intervals, each interval time duration Delta T, the loss value of the arrival time is converted into the cost by combining the arrival time and the road network charging condition, and the value function of heterogeneous travelers is established by combining the path charging;
traveler arriving at optimum time tpAnd departure time tsMake a travel time budget tbI.e. tb＝tp-ts. Wherein t isp,te,tl,tk,te,tbThe variable is a variable related to heterogeneous traveler types, so that a cost function, a probability weight function and a foreground value in the foreground value calculation process are related to a traveler type x, the unit time value of the xth traveler is marked as alpha (x), the complexity of an expression is reduced, the traveler type mark is omitted, and the cost function of a time interval path k of m on the nth day of the xth traveler is obtained:
Figure BDA0001687947200000031
in the formula ofk(n, m) is a charging value of the m-link route k on the nth day, λ1-λ6Denotes a profit bias or loss aversion coefficient, and λ when q is 1,2,5,6q<When q is 3,4, λq>0，γxRepresenting the risk factor.
The determining of the probability weight function in S3 specifically includes:
the probability weight function is as follows:
W(Pj)＝exp{-(-lnPj)θ},j＝1,2,3，4,5,6
wherein theta is a value of (0, 1)]Parameter of (2) PjThe objective probability of occurrence of each condition is expressed, and the occurrence probability is divided according to the understanding of the travel time and is converted into a standard normal distribution.
The calculation of the path foreground value comprises the following specific processes:
Figure BDA0001687947200000032
foreground value PVk(n, m) represents the foreground value of the path k for the period m on the nth day.
The departure period flow evolution model with the maximum punctual arrival probability of the S5 as the target specifically comprises the following steps:
taking m times period as research object, taking quasi-point arrival probability Z (m) as maximum target, when Z (m) is*)-Z(m)≥η1Carrying out flow transfer;
setting a rotation of a flow over a period of timeAnd (3) shifting conditions: | m*-m|·ΔT≤η2I.e. the time of adjustment of the departure time needs to be less than the threshold η2Otherwise, the traveler resists because of the risk, the flow is not transferred, and the transfer relationship of the flow between the departure time is as follows:
Figure BDA0001687947200000041
Figure BDA0001687947200000042
Figure BDA0001687947200000043
Figure BDA0001687947200000044
in the formula (I), the compound is shown in the specification,
Figure BDA0001687947200000045
the flow transfer-in-out probability, ω, of the m × time period on day n +1, respectively1The time interval transfer coefficient is expressed, and the transfer probability of the flow represented by the formula is inversely proportional to the number of transfer time intervals.
In the step S6, a path flow evolution model with the path foreground value as the maximum target is obtained,
considering the traffic transfer among paths, based on the traffic time interval transfer result, combining the foreground values of the paths in each time interval to give a path transfer probability:
Figure BDA0001687947200000046
Figure BDA0001687947200000047
Figure BDA0001687947200000048
representing the transition-in/out probability, ω, of the path k at m times of day n +12The time interval transfer coefficient is represented, and the transfer probability of the flow represented by the formula is in direct proportion to the foreground difference value;
flow rate at m times of the day n
Figure BDA0001687947200000049
Based on the path transfer probability of the flow, a daily deduction formula of the path flow is established to obtain the flow of the n +1 day m time period path k
Figure BDA00016879472000000410
Figure BDA00016879472000000411
Figure BDA00016879472000000412
The invention has the beneficial effects that:
1. the types of the travelers are divided according to different time values, so that the flow prediction of the congested toll road network is more accurate;
2. the method has the advantages that the day-to-day travel characteristics of travelers in the toll road network are considered, a transfer model of the traffic at the departure time and the travel path is established, the model can be used for predicting the traffic evolution under the real toll situation, and the reference value is provided for determining the road network toll value.
Drawings
FIG. 1 is a flow chart of the operation of the present invention.
Detailed Description
The present invention will be described in further detail with reference to examples and drawings, but the present invention is not limited to these examples.
Examples
As shown in fig. 1, a method for predicting road network traffic considering heterogeneous users includes the following steps:
and S1, updating road network parameters according to empirical learning, wherein the road network parameters comprise the average value and the variance of the running time of each path and the understanding running time of a traveler.
The path running time follows normal distribution and has Tk～N(τk,σk),τk，σkRespectively representing the running time mean and variance of the path k. The parameters of the path k in the m period of the nth day are updated by the data of the same path in the same period of the nth-1 day:
τk(n,m)＝(τk(n-1,m)·(n-1)+tk(n,m))/n
σk(n,m)2＝(σk(n-1,m)2·(n-1)+(tk(n,m)-τk(n,m))2)/n
Tkfor the running time distribution function, N denotes the normal distribution, tk(n, m) represents the actual travel time of the kth route in the period of m on the nth day, and when n is 1, tauk(1,m)＝tk free,σk(1,m)＝0,tk freeShowing the free travel time of the path.
The traveler has an estimate of the travel time of the route before traveling every day, defined as the understanding of travel time etk(n, m), updating the same path of the n-th day time period by the understanding travel time of the m-th day period path k and the actual travel time according to historical travel experience by a traveler:
Figure BDA0001687947200000051
Figure BDA0001687947200000052
is a parameter which reflects the degree of dependence of a traveler on the actual traveling time of the previous day,
Figure BDA0001687947200000053
the smaller, the greater the degree of dependence,
Figure BDA0001687947200000054
the larger, the smaller the degree of dependence,
Figure BDA0001687947200000055
the value range is [0,1 ]]When n is 1, etk(1,m)＝tk free。
S2, determining a cost function
Defining a traffic network G ═ W, A, W is a node set, A is a road segment set, road segments a ∈ A, and defining study periods of each day as [ T1, T2]]The charging period is [ T1', T2']Where T1 ≦ T1 '< T2' ≦ T2, the study period is divided into M equal time intervals, each interval being of duration Δ T. Combining the arrival time and the road network charging condition, converting the loss value of the arrival time into the cost, and combining the path charging to establish a value function of heterogeneous travelers. Traveler arriving at optimum time tpAnd departure time tsMake a travel time budget tbI.e. tb＝tp-ts. Wherein t isp,te,tl,tbThe values of the x-th class of actors in the unit time are marked as alpha (x), and the speaker type mark is omitted in order to reduce the complexity of the expression. Therefore, the value function of m time interval path k of x class travelers on the nth day is obtained:
Figure BDA0001687947200000061
in the formula ofk(n, m) is a charging value of the m-link route k on the nth day, λ1-λ6Denotes a profit bias or loss aversion coefficient, and λ when q is 1,2,5,6q<When q is 3,4, λq>0，γxRepresenting the risk factor.
S3 probability weight function determination
People have subjectivity on the knowledge of objective probability, and in order to describe the subjective process of the objective probability, a probability weight function is provided:
W(Pj)＝exp{-(-lnPj)θ},j＝1,2,3，4,5,6
theta is taken as value of (0, 1)]In the parameter between, 6 cases are given in the cost function, PjThe objective probability of occurrence of each condition is expressed, and the occurrence probability is divided according to the understanding of the travel time and is converted into a standard normal distribution.
S4, calculation of path foreground value
Figure BDA0001687947200000062
Foreground value PVk(n, m) represents the (x class traveler) foreground value of the path k at the time period m on the nth day.
S5 evolution model of departure period flow
Taking m times period as research object, taking quasi-point arrival probability Z (m) as maximum target, when Z (m) is*)-Z(m)≥η1And carrying out flow transfer. Setting the transfer condition of the flow in the time period: | m*-m|·ΔT≤η2I.e. the time of adjustment of the departure time needs to be less than the threshold η2Otherwise, the traveler resists because of the risk, and the flow does not shift. The transfer relationship of the traffic between departure times is as follows:
Figure BDA0001687947200000071
Figure BDA0001687947200000072
Figure BDA0001687947200000073
Figure BDA0001687947200000074
in the formula
Figure BDA0001687947200000075
The flow transfer-in-out probability, ω, of the m × time period on day n +1, respectively1The time interval transfer coefficient is expressed, and the transfer probability of the flow represented by the formula is inversely proportional to the number of transfer time intervals.
S6 path flow evolution model
Considering the traffic transfer among paths, based on the traffic time interval transfer result, combining the foreground values of the paths in each time interval to give a path transfer probability:
Figure BDA0001687947200000076
Figure BDA0001687947200000077
Figure BDA0001687947200000078
representing the transition-in/out probability, ω, of the path k at m times of day n +12The time interval transfer coefficient is represented, and the expression represents that the transfer probability of the flow is in direct proportion to the foreground difference value.
Flow rate at m times of the day n
Figure BDA0001687947200000079
Based on the path transfer probability of the flow, a daily deduction formula of the path flow is established to obtain the flow of the n +1 day m time period path k
Figure BDA00016879472000000710
Figure BDA00016879472000000711
Figure BDA00016879472000000712
The above embodiments are preferred embodiments of the present invention, but the present invention is not limited to the above embodiments, and any other changes, modifications, substitutions, combinations, and simplifications which do not depart from the spirit and principle of the present invention should be construed as equivalents thereof, and all such changes, modifications, substitutions, combinations, and simplifications are intended to be included in the scope of the present invention.

Claims (5)

1. A road network flow prediction method considering heterogeneous users is characterized by comprising the following steps:
s1, updating road network parameters according to empirical learning;
s2, determining a cost function in the foreground theory;
s3, determining a probability weight function;
s4, calculating a path foreground value;
s5, a departure period flow evolution model with the maximum quasi-point arrival probability as a target;
s6, obtaining a path flow evolution model with the path foreground value as the maximum target;
the determination of the cost function in the foreground theory in S2 specifically includes:
defining a traffic network G ═ W, A, W is a node set, A is a road segment set, a road segment belongs to A, each day of research time interval is defined as [ T1, T2], each day of charging time interval is defined as [ T1', T2' ], wherein T1 is not less than T1 '< T2' < T2, the research time interval is divided into M equal time intervals, each interval time duration Delta T, the loss value of the arrival time is converted into the cost by combining the arrival time and the road network charging condition, and the value function of heterogeneous travelers is established by combining the path charging;
traveler arriving at optimum time tpAnd departure time tsMake a travel time budget tbI.e. tb＝tp-tsIn, middle tp,te,tl,tk,ts,tbIs a variable relating to the type of heterogeneous traveler, tkRepresenting the actual travel time of the kth path, and hence the cost function, probability, in the process of calculating the foreground valuesThe weighting function and the foreground value are related to the type x of the traveler, the unit time value of the x-th traveler is marked as alpha (x), and the value function of the m-period path k on the nth day of the x-th traveler is as follows:
Figure FDA0003259606280000011
in the formula ofk(n, m) is a charging value of the m-link route k on the nth day, λ1-λ6Denotes a profit bias or loss aversion coefficient, and λ when q is 1,2,5,6q<When q is 3,4, λq>0，γxRepresenting a risk factor;
the departure period flow evolution model with the maximum punctual arrival probability of the S5 as the target specifically comprises the following steps:
taking m times period as research object, taking quasi-point arrival probability Z (m) as maximum target, when Z (m) is*)-Z(m)≥η1Carrying out flow transfer;
setting the transfer condition of the flow in the time period: | m*-m|·ΔT≤η2I.e. the time of adjustment of the departure time needs to be less than the threshold η2Otherwise, the traveler resists because of the risk, the flow is not transferred, and the transfer relationship of the flow between the departure time is as follows:
Figure FDA0003259606280000023
Figure FDA0003259606280000024
Figure FDA0003259606280000021
in the formula (I), the compound is shown in the specification,
Figure FDA0003259606280000025
are respectively asDay n +1 m*Time-phased flow transfer-in-out probability, omega1Expressing the time interval transfer coefficient, the transfer probability of the flow expressed by the above formula is inversely proportional to the number of transfer time intervals, fk(n+1,m*) Is represented by n +1 day m*The traffic of epoch path k.
2. The road network traffic prediction method according to claim 1, wherein the updating of road network parameters based on empirical learning is specifically:
the path running time follows normal distribution and has Tk～N(τk,σk),τk，σkRespectively representing the average value and the variance of the walking time of the path k, and updating the parameters of the path k in the m period on the nth day by the data of the same path on the (n-1) th day at the same time period:
τk(n,m)＝(τk(n-1,m)·(n-1)+tk(n,m))/n
σk(n,m)2＝(σk(n-1,m)2·(n-1)+(tk(n,m)-τk(n,m))2)/n
Tkfor the running time distribution function, N denotes the normal distribution, tk(n, m) represents the actual travel time of the kth route in the period of m on the nth day, and when n is 1, tauk(1,m)＝tk free,σk(1,m)＝0,tk freeRepresenting the free running time of the path;
the traveler has an estimate of the travel time of the route before traveling every day, defined as the understanding of travel time etk(n, m), updating the same path of the n-th day time period by the understanding travel time of the m-th day period path k and the actual travel time according to historical travel experience by a traveler:
Figure FDA0003259606280000022
Figure FDA0003259606280000031
is a parameter which reflects the degree of dependence of a traveler on the actual traveling time of the previous day,
Figure FDA0003259606280000032
the smaller, the greater the degree of dependence,
Figure FDA0003259606280000033
the larger, the smaller the degree of dependence,
Figure FDA0003259606280000034
the value range is [0,1 ]]When n is 1, etk(1,m)＝tk free。
3. The road network traffic prediction method according to claim 1, wherein the determining of the probability weight function in S3 specifically includes:
the probability weight function is as follows:
W(Pj)＝exp{-(-lnPj)θ},j＝1,2,3，4,5,6
wherein theta is a value of (0, 1)]Parameter of (2) PjThe objective probability of occurrence of each condition is expressed, and the occurrence probability is divided according to the understanding of the travel time and is converted into a standard normal distribution.
4. The road network traffic prediction method according to claim 1, wherein the calculation of the path foreground value comprises the following specific processes:
Figure FDA0003259606280000035
foreground value PVk(n, m) represents the foreground value of the path k for the period m on the nth day.
5. The road network traffic prediction method according to claim 1, wherein the path traffic evolution model with the path foreground value as the maximum target is obtained in S6,
considering the traffic transfer among paths, based on the traffic time interval transfer result, combining the foreground values of the paths in each time interval to give a path transfer probability:
Figure FDA0003259606280000036
Figure FDA0003259606280000037
Figure FDA0003259606280000038
denotes day n +1 m*Epoch Path k*Turn-in and turn-out probability of, omega2Representing a path transfer coefficient, wherein the above expression represents that the transfer probability of the flow is in direct proportion to the foreground difference value;
on the nth day m*Epoch Path k*Flow rate of
Figure FDA0003259606280000039
On the basis, a daily deduction formula of the path flow is established according to the path transfer probability of the flow to obtain n +1 day m*Epoch Path k*Flow rate of
Figure FDA00032596062800000310
Figure FDA00032596062800000311
CN201810579045.7A 2018-06-07 2018-06-07 Road network flow prediction method considering heterogeneous users Expired - Fee Related CN108847021B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810579045.7A CN108847021B (en) 2018-06-07 2018-06-07 Road network flow prediction method considering heterogeneous users

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810579045.7A CN108847021B (en) 2018-06-07 2018-06-07 Road network flow prediction method considering heterogeneous users

Publications (2)

Publication Number Publication Date
CN108847021A CN108847021A (en) 2018-11-20
CN108847021B true CN108847021B (en) 2021-12-21

Family

ID=64210573

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810579045.7A Expired - Fee Related CN108847021B (en) 2018-06-07 2018-06-07 Road network flow prediction method considering heterogeneous users

Country Status (1)

Country Link
CN (1) CN108847021B (en)

Families Citing this family (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN110211381B (en) * 2019-06-05 2021-02-09 杭州中奥科技有限公司 Traffic route distribution method and device
CN111833596B (en) * 2019-11-19 2022-04-29 东南大学 Day-by-day road section flow prediction method considering decision inertia of travelers
CN110942626B (en) * 2019-11-21 2021-09-21 华南理工大学 Road network mixed flow rate daily variation prediction method considering unmanned vehicles
CN113345252B (en) * 2021-06-08 2022-07-22 重庆大学 Short-time prediction method and device for lower-path flow of toll station

Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2010210284A (en) * 2009-03-06 2010-09-24 Denso Corp Traffic management device and traffic management method
CN107256632A (en) * 2017-08-11 2017-10-17 上海交通大学 A kind of method of traffic assignment based on the heterogeneous time value of user Yu congestion expense budget

Patent Citations (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2010210284A (en) * 2009-03-06 2010-09-24 Denso Corp Traffic management device and traffic management method
CN107256632A (en) * 2017-08-11 2017-10-17 上海交通大学 A kind of method of traffic assignment based on the heterogeneous time value of user Yu congestion expense budget

Non-Patent Citations (3)

* Cited by examiner, † Cited by third party
Title
交通事件影响下路网逐日出行动态可靠性;陈玲娟等;《交通运输系统工程与信息》;20171031;第17卷(第5期);第97-103页 *
考虑出发时刻调整的日变路网配流模型;陈玲娟等;《交通运输系统工程与信息》;20151231;第15卷(第6期);第190-196页 *
陈玲娟等.交通事件影响下路网逐日出行动态可靠性.《交通运输系统工程与信息》.2017,第17卷(第5期),第97-103页. *

Also Published As

Publication number Publication date
CN108847021A (en) 2018-11-20

Similar Documents

Publication Publication Date Title
CN108847021B (en) Road network flow prediction method considering heterogeneous users
CN101695050A (en) Dynamic load balancing method based on self-adapting prediction of network flow
Amirgholy et al. Modeling the dynamics of congestion in large urban networks using the macroscopic fundamental diagram: User equilibrium, system optimum, and pricing strategies
Christofa et al. Person-based traffic responsive signal control optimization
Qian et al. Optimal parking pricing in general networks with provision of occupancy information
Friesz et al. Dynamic congestion pricing in disequilibrium
CN104183119B (en) Based on the anti-arithmetic for real-time traffic flow distribution forecasting method pushed away of section OD
CN102137023B (en) Multicast congestion control method based on available bandwidth prediction
CN101488284A (en) Intelligent management system for road traffic condition instant prediction
CN103927887A (en) Array type FPGA traffic state prediction and control system combined with discrete speed model
CN109147324B (en) Traffic jam probability forecasting method based on user feedback mechanism
CN111695722A (en) Method for predicting short-term passenger flow in holidays of urban rail transit station
CN110942626B (en) Road network mixed flow rate daily variation prediction method considering unmanned vehicles
CN110210648A (en) Control zone strategy method for predicting based on grey shot and long term memory network
Tettamanti et al. Model predictive control in urban traffic network management
CN103870890A (en) Prediction method for traffic flow distribution of expressway network
TWI623920B (en) Speed prediction method
CN117912235B (en) Planning data processing method and system for smart city
Cenggoro et al. Dynamic bandwidth management based on traffic prediction using Deep Long Short Term Memory
Van der Zijpp et al. Estimation of origin-destination demand for dynamic assignment with simultaneous route and departure time choice
CN114495529B (en) Signal timing optimization system based on distributed model predictive control
Greguric et al. A neuro-fuzzy based approach to cooperative ramp metering
Bellemans et al. Model predictive control for ramp metering combined with extended Kalman filter-based traffic state estimation
Wang et al. Feedback route guidance applied to a large-scale express ring road
Ferrara et al. Model-based event-triggered control for freeway traffic systems

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
GR01 Patent grant
GR01 Patent grant
CF01 Termination of patent right due to non-payment of annual fee

Granted publication date: 20211221

CF01 Termination of patent right due to non-payment of annual fee