CN109190948B - Correlation analysis method for operation of large-scale aviation hub and urban traffic jam - Google Patents

Correlation analysis method for operation of large-scale aviation hub and urban traffic jam Download PDF

Info

Publication number
CN109190948B
CN109190948B CN201810950116.XA CN201810950116A CN109190948B CN 109190948 B CN109190948 B CN 109190948B CN 201810950116 A CN201810950116 A CN 201810950116A CN 109190948 B CN109190948 B CN 109190948B
Authority
CN
China
Prior art keywords
index
time period
travel time
value
airport
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.)
Active
Application number
CN201810950116.XA
Other languages
Chinese (zh)
Other versions
CN109190948A (en
Inventor
曹先彬
杜文博
张明远
汪一杰
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.)
Beihang University
Original Assignee
Beihang University
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 Beihang University filed Critical Beihang University
Priority to CN201810950116.XA priority Critical patent/CN109190948B/en
Publication of CN109190948A publication Critical patent/CN109190948A/en
Application granted granted Critical
Publication of CN109190948B publication Critical patent/CN109190948B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Images

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
    • G06N3/00—Computing arrangements based on biological models
    • G06N3/02—Neural networks
    • G06N3/04—Architecture, e.g. interconnection topology
    • G06N3/045—Combinations of networks
    • 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
    • G08G1/0133—Traffic data processing for classifying traffic situation

Landscapes

  • Engineering & Computer Science (AREA)
  • Physics & Mathematics (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • General Physics & Mathematics (AREA)
  • Theoretical Computer Science (AREA)
  • Educational Administration (AREA)
  • Chemical & Material Sciences (AREA)
  • Development Economics (AREA)
  • Economics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Analytical Chemistry (AREA)
  • Strategic Management (AREA)
  • Tourism & Hospitality (AREA)
  • Computational Linguistics (AREA)
  • Quality & Reliability (AREA)
  • Marketing (AREA)
  • General Business, Economics & Management (AREA)
  • Game Theory and Decision Science (AREA)
  • Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Artificial Intelligence (AREA)
  • Biomedical Technology (AREA)
  • Biophysics (AREA)
  • Operations Research (AREA)
  • Data Mining & Analysis (AREA)
  • Evolutionary Computation (AREA)
  • General Health & Medical Sciences (AREA)
  • Molecular Biology (AREA)
  • Computing Systems (AREA)
  • General Engineering & Computer Science (AREA)
  • Mathematical Physics (AREA)
  • Software Systems (AREA)
  • Traffic Control Systems (AREA)
  • Management, Administration, Business Operations System, And Electronic Commerce (AREA)

Abstract

The invention discloses a method for analyzing association between large-scale aviation hub operation and urban traffic jam, and belongs to the field of traffic operation conditions. Firstly, a queuing theory model is established according to the number of people who take off and output by the airplane and the number of people who input by the airplane when landing in a certain time period, the length of the team and the waiting time of the number of people are calculated, and a congestion sense index and an anxiety sense index which measure delay conditions in an airport are established. And then calculating the free flow speed and the travel time index of each road section in the road network within the range of the field radius M kilometers in the time period, calculating the average value of the travel time indexes in the time period after deleting unreasonable data, and further calculating the entropy of the travel time indexes. And finally, predicting the mean value of the travel time index and the entropy value of the travel time index in the next time period by using an LSTM neural network, and calculating the influence of the delay of the airport on the surrounding roads in the current time period. The method considers the congestion condition in the airport, quantifies the congestion condition in the airport, and is simple and feasible.

Description

Correlation analysis method for operation of large-scale aviation hub and urban traffic jam
Technical Field
The invention relates to technologies such as a queuing theory, traffic operation condition evaluation, deep learning and traffic flow prediction, in particular to a correlation analysis method for large-scale aviation hub operation and urban traffic jam.
Background
Modern transportation has been transformed from single-pavement transportation to multi-dimensional transportation systems, including pavement, underground, air and ocean, etc. Road traffic is affected not only by the road but also by the synergy of other traffic patterns in the traffic system. However, the existing analysis for predicting the traffic operation state only focuses on the mutual influence of road traffic; in the face of modern multi-element traffic systems, an analysis method for reducing the influence of the operation efficiency of a large traffic hub on surrounding roads is lacked.
Disclosure of Invention
The problem that influence of other traffic modes is not considered in the existing traffic analysis is solved. The invention provides a correlation analysis method for large-scale aviation hub operation and urban traffic jam, which introduces airport delay conditions obtained through a queuing theory as a main factor into traffic flow prediction so as to evaluate the influence of the airport delay on surrounding road traffic.
The method comprises the following specific steps:
step one, counting the number of people who should arrive in a certain time period according to a flight schedule of the certain time period, and recording the number of people who take off and output the airplane and the number of people who input the airplane when landing in the time period under the actual condition.
The time period is divided into: the 24 hours a day is divided into 48 time periods, i.e. one time period for each half hour.
The method for recording the number of the aircraft takeoff output people comprises the following steps: in each time period, the number of flights taking off at each moment is recorded, the total number m of persons which can be transported is calculated, the time interval t between the flights taking off at the next moment is recorded, and t/m is recorded. The mean and variance are calculated for all t/m in each time period.
Step two, establishing a queuing theory model M/E according to the statistical result of the step onekA/1 model for calculating the length L of the team in the time periodqWaiting time W for average personq。
The formula is as follows:
Figure BDA0001771273750000011
Figure BDA0001771273750000012
wherein
Figure BDA0001771273750000013
k is Ekλ is the demand rate, μ is the service rate.
Step three, utilizing team length LqWaiting time W for average personqEstablishing a congestion sense index α and an anxiety sense index β for measuring delay situations in an airport;
index of congestion feeling α ═ phi1ln(Lq-C)+φ2
Index of anxiety, β ═ phi3ln(Wq-K)+φ4
Wherein
Figure BDA0001771273750000023
φ1，φ2，φ3And phi4Is a normalization coefficient, and C is a congestion threshold value; k is an anxiety sensation threshold;
step four, calculating the free flow speed of each road section in the road network within the range of the field radius M kilometers in the time period;
the free flow speed is the running speed of the vehicle under the condition that the road is completely unblocked.
Step five, calculating the travel time index TTI of each road section in the time period by using the free flow speed of each road section;
the travel time index is the ratio of the time it takes for the vehicle to travel the same distance at the current speed as the free stream speed.
And step six, arranging the TTI from small to large, selecting a truncation position, deleting unreasonable data larger than the truncation position, and reserving the TTI data from zero to the truncation position for subsequent processing.
The truncation position is selected from the 99% quantile of the travel time index.
And step seven, calculating the travel time index mean value of the time slot according to the travel time indexes in the whole road network calculated in the step five, and taking the travel time index mean value as an index for evaluating the road network congestion condition.
And step eight, calculating the entropy value H (X) of the travel time index according to the discrete probability density distribution of each travel time index reserved in the step six, and using the entropy value H (X) as an index for evaluating the complexity of the road network.
The formula is as follows:
Figure BDA0001771273750000021
wherein P (TTI)i) Is the ith kind in TTI discrete probability density distributionAnd the TTI value is corresponding to the probability value, and N is the total number of TTI equal interval divisions from zero to a truncation position.
And ninthly, performing time prediction analysis by comprehensively considering characteristic quantities of airport and road indexes by using an LSTM neural network, and predicting the mean value of the travel time index and the entropy value of the travel time index of the next time period.
First, input features of the neural network include: the characteristic quantity of the airport index and the characteristic quantity of the road index passing through the connecting layer;
normalizing the number of people who should arrive in each time period, and taking the three values as the index of the airport detention condition and the characteristic quantity of the airport index through a full connecting layer, wherein the three values are the congestion index α and the anxiety index β;
the method comprises the steps of normalizing the mean value of the travel time index of each time period, normalizing the entropy value H (X) of the travel time index of each time period, normalizing the divided time periods and the daily weather condition quantized value, wherein the four values are used as road congestion condition indexes and are used as characteristic quantities of the road indexes through a full-connection layer.
The normalization of the mean value of the travel time index of each time segment specifically includes: and performing normalization processing by taking the maximum value of the travel time index mean value in each time period history as the maximum value of the historical data and taking the minimum value as the minimum value of the historical data.
The normalization of the entropy value h (x) of the travel time index for each time segment is specifically: maximum entropy value H of travel time exponential entropymaxObtained by the following formula, HmaxThe minimum value is the minimum value obtained from historical monitoring data; thereby normalizing the travel time index entropy.
The weather condition quantification of the current day is specifically as follows: the number of weather indicators is 7, etc., corresponding to a value of
Figure BDA0001771273750000022
Normalizing for the divided time segments means:the time period is changed into
Figure BDA0001771273750000031
Then, inputting the two groups of indexes passing through the full connection layer into the LSTM layer;
and finally, training the output of the LSTM layer through three full-connection layers to obtain the mean value of the travel time index of the next time period and the entropy value of the travel time index.
Step ten, calculating the influence of the delay of the airport on the surrounding roads in the current time period by using the predicted travel time index mean value and the predicted travel time index entropy of the next time period;
firstly, calculating an index k and a road complexity influence index l of influence of airport detention on road congestion;
the road congestion influence index calculation formula is as follows:
Figure BDA0001771273750000032
the road complexity influence index calculation formula is as follows:
Figure BDA0001771273750000033
wherein m (t +1) is the predicted next time period travel time index mean value, and m (t) is the current time period travel time index mean value. H (t +1) is the next time period travel time index entropy, and H (t) is the current time period travel time index entropy.
Then, k and l are normalized and used as quantitative indexes to evaluate the influence of airport delay on the congestion of surrounding roads.
Selecting the maximum value of the history and the minimum value of the history as the maximum value and the minimum value of the history, and normalizing:
Figure BDA0001771273750000034
the closer the value obtained is to 1, the greater the influence of airport delay on the surrounding roads.
Compared with the prior art, the invention has the beneficial effects that:
(1) a method for analyzing the association between the operation of a large-scale aviation hub and urban traffic jam is used for predicting the short-term traffic jam condition around an airport, not only using the previous traffic jam index, but also considering the jam condition inside the airport.
(2) A correlation analysis method for operation of a large-scale aviation hub and urban traffic jam is a method for evaluating the retention condition inside an airport and providing evaluation indexes of the jam condition in the airport, namely a jam index and an anxiety index. Through the two indexes, the congestion situation in the airport is quantified, and the influence on the road congestion around the airport is introduced.
(3) A correlation analysis method for large-scale aviation hub operation and urban traffic jam is used for evaluating the influence of airport detention on surrounding roads, and is simple and feasible.
Drawings
FIG. 1 is a flow chart of a method for analyzing the association between large-scale aviation hub operation and urban traffic congestion according to the present invention;
FIG. 2 is a LSTM neural network constructed in accordance with the present invention.
Detailed Description
The present invention will be described in further detail below with reference to the accompanying drawings.
The invention discloses a method for analyzing the association between large-scale aviation hub operation and urban traffic jam, which comprises the following steps: 1) and calculating the number of people arriving at an airport, the number of people taking off an airplane and the number of people falling the airplane by using the basic aviation timetable, and constructing a queuing theory model. 2) The crowdedness of the airport is measured by the captain and the waiting time obtained by the queuing theory model. 3) And detecting the free flow speed of the roads around the airport, and calculating the travel time index of each road. 4) And arranging the travel time indexes, selecting a truncation position, and calculating the mean value of the travel time indexes. 5) And calculating the discrete probability density of the travel time index, and solving the travel time index entropy. 6) And normalizing each required index. 7) The LSTM neural network is used for comprehensively considering the characteristics of the existing airport and road to perform time prediction analysis, after training is completed, when flight delay occurs, characteristic indexes required by the network are calculated, and the travel time mean value and the travel time entropy at the next moment are predicted; 8) and constructing an evaluation standard to quantitatively express the influence of the delay of the airport on the road congestion condition.
As shown in fig. 1, the specific steps are as follows:
step one, counting the number of people who should arrive in a certain time period according to a flight schedule of the certain time period, and recording the number of people who take off and output the airplane and the number of people who input the airplane when landing in the time period under the actual condition.
The time period is divided into: the delay condition of the airport and the road delay condition change in a cycle of one day; in order to ensure real-time performance, 24 hours a day is divided into 48 time periods, namely, each half hour is a time period, which is marked as [ h ]1,…,h47,h48]. Statistical analysis was performed.
Calculating the number of people which should arrive in a certain time period through a flight schedule to be recorded as lambda (t);
the method for recording the number of the aircraft takeoff output people comprises the following steps: in each time period, the number of flights taking off at each moment is recorded, the total number m of persons which can be transported is calculated, the time interval t between the flights taking off at the next moment is recorded, and t/m is recorded. The mean and variance, E (E), are calculated for all t/m's in each time periodK(t)) and Var (E)k(t))。
The number of people input by the airplane landing is recorded, and the total number of people landing in the time period is recorded as g (t).
Step two, establishing a queuing theory model M/E according to the statistical result of the step onekA/1 model for calculating the length L of the team in the time periodqWaiting time W for average personq。
The data used for establishing the queuing theory model are the number of people passing through the security check and the number of people taking off the airplane;
using M/EKA/1 queuing theory model. Then at hiIn the interval, the demand rate is: λ (t) ═ x (t)/0.5 human/hour
The service rate is as follows: for an Erlang distribution of order k, the distribution density is
Figure BDA0001771273750000041
The mean and variance are therefore:
Figure BDA0001771273750000042
so that there are
Figure BDA0001771273750000043
The calculation formula is as follows:
Figure BDA0001771273750000044
Figure BDA0001771273750000045
wherein
Figure BDA0001771273750000046
k is Ekλ is the demand rate, μ is the service rate.
Step three, utilizing team length LqWaiting time W for average personqEstablishing a congestion sense index α and an anxiety sense index β for measuring delay situations in an airport;
index of congestion feeling α ═ phi1ln(Lq-C)+φ2
Index of anxiety, β ═ phi3ln(Wq-K)+φ4
Wherein
Figure BDA0001771273750000051
φ1，φ2，φ3And phi4Is a normalization coefficient, and C is a congestion threshold value; k is an anxiety sensation threshold;
step four, calculating the free flow speed of each road section in the road network within the range of the field radius M kilometers in the time period;
the free flow speed is the running speed of the vehicle under the condition that the road with small traffic volume is completely unblocked.
Step five, calculating the travel time index TTI of each road section in the time period by using the free flow speed of each road section;
the travel time index is the ratio of the time it takes for the vehicle to travel the same distance at the current speed as the free stream speed.
The travel time index is calculated by the formula:
Figure BDA0001771273750000052
and step six, arranging the TTI from small to large, selecting a truncation position, deleting unreasonable data larger than the truncation position, and reserving the TTI data from zero to the truncation position for subsequent processing.
The truncation position is selected from the 99% quantile of the travel time index.
And step seven, calculating the mean value of the time period according to the travel time indexes in the whole road network calculated in the step five, and taking the mean value as an index for evaluating the road network congestion condition.
And step eight, calculating the entropy value H (X) of the travel time index according to the discrete probability density distribution of each travel time index reserved in the step six, and using the entropy value H (X) as an index for evaluating the complexity of the road network.
The formula is as follows:
Figure BDA0001771273750000053
wherein P (TTI)i) And (3) obtaining a corresponding probability value for the ith TTI in the TTI discrete probability density distribution, wherein N is the total number of TTI equal-interval divisions from zero to a truncation position.
And ninthly, performing time prediction analysis by comprehensively considering characteristic quantities of airport and road indexes by using an LSTM neural network, and predicting the mean value of the travel time index and the entropy value of the travel time index of the next time period.
Firstly, designing an LSTM neural network to carry out time sequence analysis training on traffic conditions around an airport; the model of the neural network is shown in fig. 2: the input features include: the characteristic quantity of the airport index and the characteristic quantity of the road index passing through the connecting layer;
the method specifically comprises the following steps:
the first part is airport related characteristic quantity, namely normalized values of the number of people who should arrive in each time period, a congestion index α and an anxiety index β, wherein the three values are used as indexes of airport detention conditions and are used as characteristic quantity of airport indexes through a full-connection layer;
the second part is road-related feature quantity: the method comprises the steps of normalizing the mean value of the travel time indexes of all time periods, normalizing the entropy value H (X) of the travel time index of each time period, normalizing the divided time periods and the daily weather condition quantized value, wherein the four values are used as road congestion condition indexes and are used as characteristic quantities of the road indexes through a full-connection layer.
The normalization of the mean value of the travel time index of each time segment specifically includes:
each time period history is recorded as TTImean＝mean(TTIi) And normalizing the mean value: the maximum value of the travel time index mean value is the maximum value max (TTI) of the existing datamean) The minimum value of the mean value is the minimum value min (TTI) of the existing datamean). Normalized formula is
Figure BDA0001771273750000061
The normalization of the entropy value h (x) of the travel time index for each time segment is specifically:
maximum entropy value H of travel time exponential entropymaxObtained by the following equation, the upper bound is given by the Jansen inequality, available as E (logY) < log (E (Y)), given by:
Figure BDA0001771273750000062
the lower bound of the travel time entropy is defined as the minimum value min (h (x)) of the travel time entropy of the existing data; namely, the minimum value is the minimum value obtained in the historical monitoring data; thereby carrying out exponential entropy normalization on travel timeComprises the following steps:
Figure BDA0001771273750000063
the weather condition quantification of the current day is specifically as follows: the number of weather indicators is 7, etc., corresponding to a value of
Figure BDA0001771273750000064
Normalizing for the divided time segments means: the time period is changed into
Figure BDA0001771273750000065
Then, inputting the two groups of characteristic quantity indexes passing through the full connection layer into the LSTM layer together for time series analysis;
and finally, training the output of the LSTM layer through three full-connection layers to obtain the mean value of the travel time index of the next time period and the entropy value of the travel time index.
The method specifically comprises the following steps: the LSTM network is named as a long-short term memory network, each LSTM unit cell is provided with a memory unit and a forgetting unit, and hidden Markov property can be mined and applied in operation. The traffic data has Markov characteristics and is suitable for training by using an LSTM network.
The characteristic quantities of airport and road indexes in the time period of [ t-n., t-1, t ] are used as input, and the average travel time index and the travel time index entropy in the time period of t +1 are used as label.
The method comprises the steps that a neural network is trained by utilizing the flow data of an original airport scheduled flight schedule, an actual flight schedule and roads around the airport, supervised learning is carried out by utilizing historical data through back propagation by the neural network, and a network weight table capable of accurately predicting the next time period is obtained after repeated iterative operation, so that the relation between the characteristic quantity of an airport index and the characteristic quantity of a road index and label is established. After training is finished, calculating the characteristic quantity of the airport index and the characteristic quantity of the road index by real-time detection data, and predicting the average travel time index and the travel time index entropy of the next time period through network operation; further, the influence of the delay of the airport on the surrounding roads at this time will be described.
Step ten, calculating the influence of the delay of the airport on the surrounding roads in the current time period by using the predicted travel time index mean value and the predicted travel time index entropy of the next time period;
firstly, calculating an index k and a road complexity influence index l of influence of airport detention on road congestion;
the road congestion influence index calculation formula is as follows:
Figure BDA0001771273750000071
the road complexity influence index calculation formula is as follows:
Figure BDA0001771273750000072
wherein m (t +1) is the predicted next time period travel time index mean value, and m (t) is the current time period travel time index mean value. H (t +1) is the next time period travel time index entropy, and H (t) is the current time period travel time index entropy.
Then, k and l are normalized and used as quantitative indexes to evaluate the influence of airport delay on the congestion of surrounding roads.
Selecting the maximum value of the history and the minimum value of the history as the maximum value and the minimum value of the history, and normalizing:
Figure BDA0001771273750000073
the closer the value obtained is to 1, the greater the influence of airport delay on the surrounding roads.
The above embodiments are only examples of the present invention, and are not intended to limit the present invention, and the scope of the present invention is defined by the claims. Various modifications and equivalents may be made by those skilled in the art within the spirit and scope of the present invention, and such modifications and equivalents should also be considered as falling within the scope of the present invention.

Claims (5)

1. A method for analyzing association between large-scale aviation hub operation and urban traffic jam is characterized by comprising the following specific steps:
step one, counting the number of people who should arrive in a certain time period according to a flight schedule of the certain time period, and recording the number of people who take off and output the airplane and the number of people who input the airplane when landing in the time period under the actual condition;
step two, establishing a queuing theory model M/E according to the statistical result of the step onekA/1 model for calculating the length L of the team in the time periodqWaiting time W for average personq；
The formula is as follows:
Figure FDA0002433516410000011
Figure FDA0002433516410000012
wherein
Figure FDA0002433516410000013
k is Ekλ is demand rate, μ is service rate;
step three, utilizing team length LqWaiting time W for average personqEstablishing a congestion sense index α and an anxiety sense index β for measuring delay situations in an airport;
index of congestion feeling α ═ phi1ln(Lq-C)+φ2
Index of anxiety, β ═ phi3ln(Wq-K)+φ4
Wherein
Figure FDA0002433516410000014
φ1，φ2，φ3And phi4Is a normalization coefficient, and C is a congestion threshold value; k is an anxiety sensation threshold;
step four, calculating the free flow speed of each road section in the road network within the range of the field radius M kilometers in the time period;
step five, calculating the travel time index TTI of each road section in the time period by using the free flow speed of each road section;
step six, arranging the TTI from small to large, selecting a truncation position, deleting unreasonable data larger than the truncation position, and reserving the TTI data from zero to the truncation position for subsequent processing;
step seven, calculating the travel time index mean value of the time slot according to the travel time indexes in the whole road network calculated in the step five, and taking the travel time index mean value as an index for evaluating the road network congestion condition;
step eight, calculating entropy values H (X) of the travel time indexes according to the discrete probability density distribution of each travel time index reserved in the step six, and using the entropy values H (X) as indexes for evaluating the complexity degree of the road network;
the formula is as follows:
Figure FDA0002433516410000015
wherein P (TTI)i) Obtaining a probability value corresponding to the ith TTI value in the TTI discrete probability density distribution, wherein N is the TTI equally-spaced dividing total number from zero to a truncation position;
step nine, performing time prediction analysis by comprehensively considering characteristic quantities of airport and road indexes by using an LSTM neural network, and predicting the mean value of the travel time index and the entropy value of the travel time index of the next time period;
first, input features of the neural network include: the characteristic quantity of the airport index and the characteristic quantity of the road index passing through the connecting layer;
normalizing the number of people who should arrive in each time period, and taking the three values as the index of the airport detention condition and the characteristic quantity of the airport index through a full connecting layer, wherein the three values are the congestion index α and the anxiety index β;
normalizing the mean value of the travel time index of each time period, normalizing the entropy value H (X) of the travel time index of each time period, normalizing the divided time periods and the daily weather condition quantized value, wherein the four values are used as road congestion condition indexes and are used as characteristic quantities of the road indexes through a full connection layer;
the normalization of the mean value of the travel time index of each time segment specifically includes: the maximum value of the travel time index mean value in each time period history is the maximum value of the historical data, and the minimum value is the minimum value of the historical data, so that normalization processing is carried out;
the normalization of the entropy value h (x) of the travel time index for each time segment is specifically: maximum entropy value H of travel time exponential entropymaxObtained by the following formula, HmaxThe minimum value is the minimum value obtained from historical monitoring data; thereby normalizing the travel time index entropy;
the weather condition quantification of the current day is specifically as follows: the number of weather indicators is 7, etc., corresponding to a value of
Figure FDA0002433516410000021
Normalizing for the divided time segments means: the time period is changed into
Figure FDA0002433516410000022
Then, inputting the two groups of indexes passing through the full connection layer into the LSTM layer;
finally, training the output of the LSTM layer through three full-connection layers to obtain the mean value of the travel time index of the next time period and the entropy value of the travel time index;
step ten, calculating the influence of the delay of the airport on the surrounding roads in the current time period by using the predicted travel time index mean value and the predicted travel time index entropy of the next time period;
the method specifically comprises the following steps:
firstly, calculating an index k and a road complexity influence index l of influence of airport detention on road congestion;
the road congestion influence index calculation formula is as follows:
Figure FDA0002433516410000023
the road complexity influence index calculation formula is as follows:
Figure FDA0002433516410000024
wherein m (t +1) is the predicted next time period travel time index mean value, and m (t) is the current time period travel time index mean value; h (t +1) is the travel time index entropy of the next time period, and H (t) is the travel time index entropy of the current time period;
then, normalizing k and l to be used as a quantitative index to evaluate the influence of airport delay on the congestion of surrounding roads;
selecting the maximum value of the history and the minimum value of the history as the maximum value and the minimum value of the history, and normalizing:
Figure FDA0002433516410000025
the closer the value obtained is to 1, the greater the influence of airport delay on the surrounding roads.
2. The method for analyzing the association between the operation of the large-scale aviation hub and the urban traffic jam as claimed in claim 1, wherein in the step one, the time period is divided into: dividing 24 hours a day into 48 time periods, namely, each half hour is a time period;
the method for recording the number of the aircraft takeoff output people comprises the following steps: in each time period, recording the number of flights taking off at each moment, calculating the total number m of persons transportable by the flights, recording the time interval t between the flights taking off at the next moment, and recording t/m; the mean and variance are calculated for all t/m in each time period.
3. The method as claimed in claim 1, wherein the free stream speed in step four is the driving speed of the vehicle when the road is completely clear.
4. The method as claimed in claim 1, wherein the travel time index in step five is a ratio of time taken for the vehicle to travel at the same distance at the current speed and the free stream speed.
5. The method as claimed in claim 1, wherein the intercepting location in step six is selected from 99% quantiles of travel time index.
CN201810950116.XA 2018-08-20 2018-08-20 Correlation analysis method for operation of large-scale aviation hub and urban traffic jam Active CN109190948B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201810950116.XA CN109190948B (en) 2018-08-20 2018-08-20 Correlation analysis method for operation of large-scale aviation hub and urban traffic jam

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201810950116.XA CN109190948B (en) 2018-08-20 2018-08-20 Correlation analysis method for operation of large-scale aviation hub and urban traffic jam

Publications (2)

Publication Number Publication Date
CN109190948A CN109190948A (en) 2019-01-11
CN109190948B true CN109190948B (en) 2020-06-12

Family

ID=64918512

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201810950116.XA Active CN109190948B (en) 2018-08-20 2018-08-20 Correlation analysis method for operation of large-scale aviation hub and urban traffic jam

Country Status (1)

Country Link
CN (1) CN109190948B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN112967013B (en) * 2021-02-20 2024-03-19 海南太美航空股份有限公司 Method, system and electronic device for determining take-off time of pre-opened flight
CN113779761B (en) * 2021-08-10 2023-11-24 南京莱斯信息技术股份有限公司 Crowd air defense organization data analysis system and method
CN113570867B (en) * 2021-09-26 2021-12-07 西南交通大学 Urban traffic state prediction method, device, equipment and readable storage medium

Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2009259158A (en) * 2008-04-21 2009-11-05 Toyota Central R&D Labs Inc Traffic state simulation device, and program
CN103413263A (en) * 2013-08-28 2013-11-27 北京交通发展研究中心 Travel time index entropy traffic circulation evaluation method
CN105069715A (en) * 2015-07-17 2015-11-18 四川精工伟达智能技术股份有限公司 Smart tourism management system and management method based on the system
CN105389980A (en) * 2015-11-09 2016-03-09 上海交通大学 Short-time traffic flow prediction method based on long-time and short-time memory recurrent neural network
CN106652441A (en) * 2015-11-02 2017-05-10 杭州师范大学 Urban road traffic condition prediction method based on spatial-temporal data

Family Cites Families (2)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN108629973A (en) * 2018-05-11 2018-10-09 四川九洲视讯科技有限责任公司 Road section traffic volume congestion index computational methods based on fixed test equipment
CN108898838B (en) * 2018-08-03 2020-08-28 首都经济贸易大学 Method and device for predicting airport traffic jam based on LSTM model

Patent Citations (5)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
JP2009259158A (en) * 2008-04-21 2009-11-05 Toyota Central R&D Labs Inc Traffic state simulation device, and program
CN103413263A (en) * 2013-08-28 2013-11-27 北京交通发展研究中心 Travel time index entropy traffic circulation evaluation method
CN105069715A (en) * 2015-07-17 2015-11-18 四川精工伟达智能技术股份有限公司 Smart tourism management system and management method based on the system
CN106652441A (en) * 2015-11-02 2017-05-10 杭州师范大学 Urban road traffic condition prediction method based on spatial-temporal data
CN105389980A (en) * 2015-11-09 2016-03-09 上海交通大学 Short-time traffic flow prediction method based on long-time and short-time memory recurrent neural network

Also Published As

Publication number Publication date
CN109190948A (en) 2019-01-11

Similar Documents

Publication Publication Date Title
Yazdi et al. Flight delay prediction based on deep learning and Levenberg-Marquart algorithm
CN103530704B (en) A kind of air dynamic traffic volume in terminal airspace prognoses system and method thereof
CN115240431B (en) Real-time online simulation system and method for traffic flow of highway toll station
CN109190948B (en) Correlation analysis method for operation of large-scale aviation hub and urban traffic jam
Li et al. Identifying important variables for predicting travel time of freeway with non-recurrent congestion with neural networks
CN109522372A (en) The prediction technique of civil aviaton field passenger value
CN102694800A (en) Gaussian process regression method for predicting network security situation
CN104732286A (en) Method for sensing safety situation of urban rail transit station
CN115359659B (en) Lane opening and closing configuration method and system
CN105160201A (en) Genetic algorithm back propagation (GABP) neural network based controller workload prediction method and system
CN106779222B (en) Airport ground waiting time prediction method and device
CN112270445B (en) Flight delay wave and comprehensive evaluation method based on statistical analysis and classification prediction
CN114943356A (en) Airport arrival passenger taxi short-time demand integrated prediction method
Zhang et al. Data‐driven flight time prediction for arrival aircraft within the terminal area
CN110796315B (en) Departure flight delay prediction method based on aging information and deep learning
CN110443448A (en) A kind of aircraft seat in the plane classification prediction technique and system based on two-way LSTM
CN115953186A (en) Network appointment demand pattern recognition and short-time demand prediction method
CN115269758A (en) Passenger-guidance-oriented road network passenger flow state deduction method and system
CN109523082B (en) CNN-LSTM flight normal clearance rate prediction method
Liu et al. Understanding urban bus travel time: Statistical analysis and a deep learning prediction
CN109409563B (en) Method, system and storage medium for analyzing real-time number of people in public transport operation vehicle
Mott et al. Estimation of aircraft operations at airports using nontraditional statistical approaches
Zhou et al. The landing safety prediction model by integrating pattern recognition and Markov chain with flight data
CN117669837A (en) Airport land side transfer traffic mode selection prediction method and system
CN115796030A (en) Traffic flow prediction method based on graph convolution

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