An airport cluster flight collaborative release sequencing method

Abstract

The application relates to flight release, in particular to an airport cluster flight cooperative release sequencing method, which takes the total delay time and total delay cost minimization of flights in an airport cluster as double optimization targets, determines constraint conditions, establishes a cooperative release model, obtains flight plans of all departure airports, pre-processes the flight plans to serve as inputs of the cooperative release model, solves the optimal flight sequencing by using a genetic algorithm, and realizes airport cluster flight cooperative release sequencing; the technical scheme can effectively overcome the defect that the solving result cannot provide auxiliary decision for current airport cluster flight cooperative release in the prior art.

Description

Technical Field

[0001] This invention relates to flight release, specifically to a method for coordinated flight release sequencing in an airport cluster. Background Technology

[0002] With the gradual maturation of airport clusters in China, accelerating the construction of world-class airport clusters such as the Beijing-Tianjin-Hebei region, the Yangtze River Delta, and the Guangdong-Hong Kong-Macau Greater Bay Area has become an important task for my country's civil aviation in recent years. Due to the geographical proximity of airports within these clusters, scheduling adjustments and coordinated flight releases between them present significant challenges. Therefore, research on coordinated flight releases within airport clusters is of great practical significance, and the construction of relevant platforms for coordinated flight releases within airport clusters is urgently needed.

[0003] Current research on the coordinated release of flights in airport clusters lacks a comprehensive consideration of multiple optimization objectives, and the constraints involved are mostly theoretical, making it impossible for the solutions to provide auxiliary decision-making for the coordinated release of flights in airport clusters. Summary of the Invention

[0004] (a) Technical problems to be solved

[0005] In view of the above-mentioned shortcomings of the existing technology, the present invention provides a method for coordinated release and sequencing of flights in airport clusters, which can effectively overcome the deficiency that the solution results of the existing technology cannot provide auxiliary decision-making for the coordinated release of flights in the current airport clusters.

[0006] (II) Technical Solution

[0007] To achieve the above objectives, the present invention provides the following technical solution:

[0008] A method for coordinated flight release and sequencing in an airport cluster includes the following steps:

[0009] S1. The dual optimization objectives are to minimize the total flight delay time and total flight delay cost of the airport cluster, and the constraints are determined to establish a collaborative release model.

[0010] S2. Obtain flight schedules for all departing airports, preprocess them, and use them as input for the collaborative release model;

[0011] S3. Use a genetic algorithm to solve for the optimal flight order and realize the coordinated release order of flights in the airport group.

[0012] Preferably, the total flight delay time y(t) of the airport group is calculated using the following formula:

[0013]

[0014] Where M is the airport group set, m is the airport in the airport group set, and Fm For airport m, the collection of departing flights within the next time period T, f im Let t be the i-th departing flight in airport m, and t be the departure time. Let i be the departure time of the i-th departing flight in airport m. Let be the estimated departure time of the i-th departing flight in airport m;

[0015] Estimated departure time The estimated time to remove wheel chocks is the sum of the airport's average taxiing time, which is the average taxiing time of all departing flights within the selected time period of airport m.

[0016] The total flight delay cost c(t) of the airport group is calculated using the following formula:

[0017]

[0018] Among them, C h This represents the average delay cost per flight for airline h.

[0019] Preferably, the total flight delay cost of the airport cluster includes ground delay costs and passenger delay costs, and the ground delay cost a for a single flight per unit time is denoted as 'a'. g Calculate using the following formula:

[0020]

[0021] Among them, O c The annual operating cost of the airline is N, the number of flights is N, and t1 is the average operating time of each flight, with the average operating time t1 of each flight uniformly set to 2 hours.

[0022] Passenger delay cost per unit time V p Calculate using the following formula:

[0023] V p =η·W avg / t2

[0024] Where η is the time value coefficient determined based on the differences in travelers' travel purposes, and W avg t2 is the average annual wage of employees nationwide, and t2 is the average effective working hours in one year, t2 = 2000;

[0025] Passenger delay cost β per unit time for a single flight t Calculate using the following formula:

[0026] β t =N'·ω·V p

[0027] Where N' is the number of seats on the flight, and ω is the load factor of the flight.

[0028] Preferably, determining the constraints in S1 includes:

[0029] Using runway clearance as a constraint, two consecutively taking off flights must meet the wake turbulence separation standard, expressed by the following formula:

[0030] t f' ≥t f +σ

[0031] Among them, t f For the departure time of flight f, t f' Let σ be the departure time of flight f' following flight f, and σ be the wake turbulence interval to be maintained.

[0032] Preferably, determining the constraints in S1 includes:

[0033] Using runway dynamic hourly capacity as a constraint, the hourly time slots to which all flights belong are obtained. Based on the historical flight schedules of peak months, the runway departure traffic for each hourly time slot is statistically analyzed. For any given hourly time slot, the statistically analyzed runway departure traffic for that hourly time slot is arranged sequentially, and a suitable runway departure traffic is selected as the runway capacity for that hourly time slot. The runway capacity results corresponding to all hourly time slots are taken as the dynamic hourly runway capacity, satisfying the following formula:

[0034]

[0035] Where M is the airport group set, m is the number of airports in the airport group set, and p mt' Let D be the runway departure flow of airport m during time period t'. mt' Let m be the runway capacity of airport m during time period t'.

[0036] Preferably, determining the constraints in S1 includes:

[0037] Using the sector's dynamic hourly capacity as a constraint, and based on historical flight schedules for peak months, flights are matched to corresponding hourly time slots for departure traffic aggregation. For any given hourly time slot, the departure traffic within that time slot is sequentially arranged, and a suitable departure traffic volume is selected as the departure capacity for that hourly time slot. The sum of traffic flows for all air routes within a sector does not exceed the sector's departure capacity, expressed by the following formula:

[0038]

[0039] Where K is the number of routes within the sector, and n j For the traffic flow of each air route in the future time period T, Let S be the off-site capacity of sector S in the future time period T.

[0040] Preferably, determining the constraints in S1 includes:

[0041] Using the route through-point interval under flow control as a constraint, the through-point interval of some routes will be restricted to a certain extent due to flow control, that is, a certain time interval needs to be maintained between adjacent flights, which is expressed by the following formula:

[0042]

[0043] in, Let i be the departure time of the i-th and i+1-th departing flights in airport m. These represent the arrival and departure points p of the i-th and (i+1)-th departing flights in airport m, respectively. l Flight time, Ωp l (t”) represents the exit point p. l The release interval during the "t" period.

[0044] Preferably, determining the constraints in S1 includes:

[0045] With the constraint that the order of flights passing through the same departure point remains unchanged, the order of flights from different airports must remain consistent, as expressed by the following formula:

[0046]

[0047] in, For passing through departure point p l The order of the k-th flight in the original set of flight orders. For passing through departure point p l The order of the k-th flight in the latest set of flight sequences.

[0048] Preferably, determining the constraints in S1 includes:

[0049] Using the maximum flight sequence offset as a constraint, to prevent excessive flight adjustments from affecting the normal operation of air traffic control, it is necessary to limit the maximum offset of a single flight sequence, expressed by the following formula:

[0050] 0≤δ≤20

[0051] Where δ is the absolute value of the difference between the flight's index in the original departure sequence set and its index in the latest departure sequence set.

[0052] Preferably, in S3, a genetic algorithm is used to solve for the optimal flight order, realizing the coordinated release order of flights in the airport group, including:

[0053] S31. Encode the departure delay time series of all departing flights during peak hours in the airport cluster, and treat each feasible departure delay time series as a chromosome, i.e., a population individual.

[0054] S32. Normalize the dual optimization objectives consisting of total flight delay time and total flight delay cost, and construct a fitness function that meets the release objective by setting a proportional coefficient to reflect the preference of the optimization scheme for reducing delay time and reducing delay cost.

[0055] S33. Select individuals with larger fitness function values ​​from the population as the solution set, and perform crossover recombination on some chromosomes of two parent individuals in the solution set with a certain probability to form new population individuals.

[0056] S34. Perform a single-point mutation on a certain departure delay time in the departure delay time series in the solution set with a certain probability;

[0057] S35, proceed to S33, until the termination condition of the genetic algorithm is met, and output the optimal flight order;

[0058] The fitness function F(t) is expressed by the following formula:

[0059]

[0060] Where y(t) is the total flight delay time of the airport group, c(t) is the total flight delay cost of the airport group, and ω1 and ω2 are the ratio coefficients of the total flight delay time and the total flight delay cost, respectively.

[0061] (III) Beneficial Effects

[0062] Compared with existing technologies, the airport cluster flight coordinated release sequencing method provided by this invention can optimize both the operational efficiency and operating cost of airport cluster flights. At the same time, it determines constraints that fully combine actual operating conditions, such as runway release interval, runway dynamic hourly capacity, sector dynamic hourly capacity, route passing point interval under flow control, unchanged passing point order, and maximum flight order offset. This enables the output optimal flight sequence to provide auxiliary decision-making for the current airport cluster flight coordinated release, and promotes the establishment of a good flight coordinated release order in the airport cluster. Attached Figure Description

[0063] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the accompanying drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are merely some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without any creative effort.

[0064] Figure 1 This is a schematic diagram of the process of the present invention;

[0065] Figure 2 This is a schematic diagram of the process of using a genetic algorithm to obtain the optimal flight order in this invention. Detailed Implementation

[0066] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without creative effort are within the scope of protection of the present invention.

[0067] A method for coordinated flight release and sequencing in an airport cluster, such as Figure 1 As shown, S1 takes minimizing the total flight delay time and total flight delay cost of the airport cluster as the dual optimization objectives, and determines the constraints to establish a collaborative release model;

[0068] S2. Obtain flight schedules for all departing airports, preprocess them, and use them as input for the collaborative release model;

[0069] S3. Use a genetic algorithm to solve for the optimal flight order and realize the coordinated release order of flights in the airport group.

[0070] In this technical solution, considering both the operational efficiency and cost of airport cluster flights, a dual optimization objective is set to maximize the interests of all parties, balancing the efficiency and economic cost of the coordinated release strategy. The preference of the coordinated release strategy for reducing delay time and costs is reflected by adjusting the weight of the dual optimization objective in the release objective. The dual optimization objectives include:

[0071] ① The total flight delay time y(t) of the airport cluster is calculated using the following formula:

[0072]

[0073] Where M is the airport group set, m is the airport in the airport group set, and F m For airport m, the collection of departing flights within the next time period T, f im Let t be the i-th departing flight in airport m, and t be the departure time. Let i be the departure time of the i-th departing flight in airport m. Let be the estimated departure time of the i-th departing flight in airport m;

[0074] Estimated departure time The estimated time to remove wheel chocks is the sum of the airport's average taxiing time, which is the average taxiing time of all departing flights within the selected time period of airport m.

[0075] ② The total flight delay cost c(t) of the airport cluster is calculated using the following formula:

[0076]

[0077] Among them, C h This represents the average delay cost per flight for airline h.

[0078] The above technical solution includes both ground delay costs and passenger delay costs for the total flight delay cost of the airport cluster.

[0079] 1) Ground delay costs refer to the costs incurred due to waiting on the ground. These losses are directly related to flight operating costs, which include fuel costs, takeoff and landing and parking fees, depreciation costs, aircraft maintenance and repair costs, employee salary costs, etc.

[0080] Ground delay cost per unit time for a single flight g Calculate using the following formula:

[0081]

[0082] Among them, O c The annual operating cost of the airline is N, the number of flights is N, and t1 is the average operating time of each flight, with the average operating time t1 of each flight uniformly set to 2 hours.

[0083] By extracting financial data from major domestic airlines such as Air China, China Eastern, China Southern, Hainan Airlines, Shandong Airlines, Spring Airlines, Juneyao Airlines, and China Express Airlines, and calculating the average costs of each airline, the ground delay cost per unit time for a single flight (yuan / hour) can be obtained, as shown in Table 1:

[0084] Table 1. Ground delay costs per unit time for a single flight of major domestic airlines.

[0085]

[0086]

[0087] 2) Passenger delay cost refers to the time passengers spend due to flight delays, preventing them from creating more value. The passenger delay cost V per unit of time is... p Calculate using the following formula:

[0088] V p =η·W avg / t2

[0089] Wherein, η is the time value coefficient determined based on the differences in passengers' travel purposes. η is a key parameter that determines the magnitude of passengers' time value. According to data from the CAPSE survey, civil aviation passengers' travel purposes are mainly divided into business travel, tourism and leisure, and visiting relatives and friends. The proportions of these three types of travel purposes in the entire civil aviation industry are 3.7:2:1. According to relevant literature research, the proportions of passengers' time value corresponding to these three types of travel purposes are 14:5:5.

[0090] W avg t2 is the average annual wage of employees nationwide (according to data from the National Bureau of Statistics, the average annual wage of employees nationwide in 2021 was 88,115 yuan), and t2 is the average effective working hours in one year, t2 = 2000;

[0091] Passenger delay cost β per unit time for a single flight t Calculate using the following formula:

[0092] β t =N'·ω·V p

[0093] Where N' is the number of seats on the flight, and ω is the load factor of the flight.

[0094] The constraints in the technical solution of this application include:

[0095] ① Using runway clearance as a constraint, two consecutively taking off flights must meet the wake turbulence separation standard, expressed by the following formula:

[0096] t f' ≥t f +σ

[0097] Among them, t f For the departure time of flight f, t f' Let f be the departure time of flight f' following flight f, and σ be the wake interval to be maintained.

[0098] For runway separation constraints, airports implementing the latest wake separation standard (RECAT) should follow the RECAT standards, as shown in Tables 2 and 3:

[0099] Table 2 RECAT's latest classification of models

[0100]

[0101]

[0102] Table 3. RECAT-specified wake interval standards (km)

[0103]

[0104] Airports that have not yet implemented the latest wake separation standard (RECAT) still follow the provisions of CCAR-93-R5, as shown in Table 4:

[0105] Table 4. CCAR-93-R5 Radar Wake Separation Standards (km)

[0106]

[0107] In Table 4, MRS represents the minimum radar wake separation, which is 6 km; J represents Airbus A380-800.

[0108] ② Using runway dynamic hourly capacity as a constraint, obtain the hourly time slots to which all flights belong. Based on the historical flight schedules of the peak month, calculate the runway departure traffic for each hourly time slot. For any given hourly time slot, rank the runway departure traffic for that hourly time slot sequentially, select a suitable runway departure traffic (selecting the runway departure traffic at the 95th quantile) as the runway capacity for that hourly time slot, and use the runway capacity results corresponding to all hourly time slots as the dynamic hourly runway capacity, satisfying the following formula:

[0109]

[0110] Where M is the airport group set, m is the number of airports in the airport group set, and p mt' Let D be the runway departure flow of airport m during time period t'. mt' Let m be the runway capacity of airport m during time period t';

[0111] Compared to the traditional static departure capacity, the dynamic hourly capacity constraint for runways is characterized by dividing the time period into hourly units and evaluating the runway capacity for each hourly period. The runway capacity corresponding to different hourly periods can reflect the congestion and delay levels of the system in different time periods.

[0112] ③ Using the sector's dynamic hourly capacity as a constraint, based on the historical flight schedule of the peak month, flights are matched to the corresponding hourly time slots for departure traffic aggregation. For any given hourly time slot, the departure traffic within that time slot is sequentially arranged, and a suitable departure traffic (selecting the departure traffic at the 95th quantile) is taken as the departure capacity for that hourly time slot. The sum of traffic traffic on all routes within the sector does not exceed the sector's departure capacity, expressed by the following formula:

[0113]

[0114] Where K is the number of routes within the sector, and n j For the traffic flow of each air route in the future time period T, Let S be the off-site capacity of sector S in the future time period T;

[0115] The sector dynamic hourly capacity constraint is similar to the runway dynamic hourly capacity constraint. The sector dynamic hourly capacity also highlights the congestion level and delay level of the sector at different times.

[0116] ④ Using the route through-point interval under flow control as a constraint, due to the influence of flow control, the through-point interval of some routes will be restricted to a certain extent, that is, a certain time interval needs to be maintained between adjacent flights, expressed by the following formula:

[0117]

[0118] in, Let i be the departure time of the i-th and i+1-th departing flights in airport m. These represent the arrival and departure points p of the i-th and (i+1)-th departing flights in airport m, respectively. l Flight time, Ωp l (t”) represents the exit point p. l The release interval during the "t" period;

[0119] ⑤ Using the unchanged departure order as a constraint, in order to ensure the orderly operation of air traffic control, the order of flights passing through the same departure point from different airports needs to remain consistent, expressed by the following formula:

[0120]

[0121] in, For passing through departure point p l The order of the k-th flight in the original set of flight orders. For passing through departure point p l The order of the k-th flight in the latest set of flight sequences;

[0122] ⑥ Using the maximum flight sequence offset as a constraint, in order to avoid excessive flight adjustments affecting the normal operation of air traffic control, it is necessary to limit the maximum offset of a single flight sequence, expressed by the following formula:

[0123] 0≤δ≤20

[0124] Where δ is the absolute value of the difference between the flight's index in the original departure sequence set and its index in the latest departure sequence set.

[0125] In this application's technical solution, flight schedules from all departing airports are obtained, preprocessed, and then used as input to the collaborative release model, including:

[0126] The system obtains pre-allocation information of all flight takeoff runways from the departure airport, matches departure point information according to airport operating regulations and flight takeoff runway allocation, and calculates the average flight time from each airport to the departure point based on historical ADSB data.

[0127] Through the above process, flight plans for all departure airports can be obtained. The flight plans include information such as airport, flight number, estimated wheel chock removal time, aircraft type (super heavy J, heavy B, general heavy C, medium M, and light L), matching takeoff runway number, departure point used, average flight time from the airport to the departure point, and the airline.

[0128] In the technical solution of this application, such as Figure 2 As shown, the optimal flight order is obtained using a genetic algorithm to achieve coordinated flight release ordering for airport clusters, including:

[0129] S31. Encode the departure delay time series of all departing flights during peak hours of the airport cluster, and treat each feasible departure delay time series as a chromosome, i.e., a population individual (in order to expand the solution search space of the problem, the population size of the problem is set to 250).

[0130] S32. Normalize the dual optimization objectives consisting of total flight delay time and total flight delay cost, and construct a fitness function that meets the release objective by setting a proportional coefficient to reflect the preference of the optimization scheme for reducing delay time and reducing delay cost.

[0131] S33. Select individuals with larger fitness function values ​​from the population as the solution set, and perform crossover recombination on some chromosomes of two parent individuals in the solution set with a certain probability to form new population individuals.

[0132] S34. Perform a single-point mutation on a certain departure delay time in the departure delay time series in the solution set with a certain probability;

[0133] S35, proceed to S33, until the termination condition of the genetic algorithm is met (the genetic algorithm terminates at generation 8000), and output the optimal flight sort.

[0134] Fitness functions are generally derived from objective functions. Based on the dual optimization objective of cooperative release, a corresponding fitness function is constructed. The fitness function value is used to measure the excellence of the cooperative release strategy represented by each chromosome. The fitness function F(t) is expressed as follows:

[0135]

[0136] Where y(t) is the total flight delay time of the airport group, c(t) is the total flight delay cost of the airport group, and ω1 and ω2 are the ratio coefficients of the total flight delay time and the total flight delay cost, respectively.

[0137] Individuals with larger fitness function values ​​(higher fitness) are more likely to be selected and passed on to the next generation. The cooperative release strategy represented by the optimal middle population individual obtained after G generations of inheritance is determined as the airport group flight cooperative release strategy.

[0138] In this embodiment, the busiest airport cluster in China, the Greater Bay Area, was selected as the research object, comprising five inland airports: Guangzhou, Shenzhen, Zhuhai, Huizhou, and Foshan. Based on historical data, a day with relatively high flight volume and classic flow control was chosen, namely May 8, 2021. Flight schedule data from 6:00 AM to 10:00 AM (morning peak) was selected, as shown in Table 5.

[0139] Table 5 Flight Schedule Data

[0140]

[0141] In addition, to fully validate the collaborative release model, flow control regarding the departure point LMN and YIN during this time period was added as one of the constraints for optimizing the collaborative release strategy.

[0142] ① Runway release interval constraint parameter settings

[0143] Based on air traffic control experience, using an aircraft speed of 180 knots, the distance interval between aircraft is converted into wake turbulence interval. The results of the RECAT wake turbulence distance interval to wake turbulence interval conversion are shown in Table 6.

[0144] Table 6. Converting RECAT wake distance intervals to wake time intervals (minutes)

[0145]

[0146] The results of converting CCAR-93-R5 wake distance intervals to wake time intervals are shown in Table 7:

[0147] Table 7. CCAR-93-R5 wake distance interval converted to wake time interval (minutes)

[0148]

[0149]

[0150] ② Runway dynamic hourly capacity constraint parameter settings

[0151] Based on historical data from April to May 2021, the dynamic hourly capacity of runways for each airport and runway was statistically analyzed. Taking Guangzhou Airport's Runway 19 as an example, the dynamic hourly capacity values ​​for the selected time period from 6:00 to 10:59 are shown in Table 8:

[0152] Table 8. Dynamic Hourly Capacity of Runway 19 at Guangzhou Airport

[0153] Hour Runway dynamic hourly capacity 6 18 7 28 8 28 9 26 10 28

[0154] The data used in the table above is historical data from May 8, 2021. Guangzhou Baiyun Airport uses runways 19 and 20R, both of which are primary departure runways; Shenzhen Bao'an Airport uses runways 15 and 16, with runway 15 being the primary departure runway; Zhuhai Jinwan Airport uses runway 23.

[0155] ③ Setting route transit interval constraint parameters under flow control conditions

[0156] Based on the actual situation of the selected time period, corresponding flow control measures were chosen to enrich the scenario. Assuming the time interval between passing the LMN point is 7 minutes and the time interval between passing the YIN point is 5 minutes, the route passing interval is predicted based on the calculated takeoff time (CTOT) of the flights. The average flight time from different airports to the departure point is calculated using historical ADSB data, as shown in Table 9.

[0157] Table 9. Average flight time from different airports to departure points.

[0158]

[0159]

[0160] ④ Sector dynamic hourly capacity constraint parameter settings

[0161] Based on the sector used by the departing flight, taking Guangzhou 01 sector as an example, the dynamic hourly capacity values ​​of the sector during the selected time period from 6:00 to 10:59 are shown in Table 10:

[0162] Table 10: Dynamic Hourly Capacity Values ​​of Sector 01 in Guangzhou

[0163] Hour Sector dynamic hourly capacity 6 20 7 24 8 24 9 18 10 24

[0164] ⑤ Setting the maximum offset constraint parameter for flight sequence

[0165] To prevent flight reversals or excessive flight sequence adjustments from affecting the normal operation of air traffic control, the maximum flight sequence offset is limited to 20.

[0166] A genetic algorithm program for the coordinated flight release problem of airport clusters was written in Python 4.0. The genetic parameters in the genetic algorithm were set as follows: population size 250, maximum number of generations 8000, and crossover probability set to 0.2 when the number of generations is less than 2000, 0.3 when the number of generations is between 2100 and 5000, and 0.4 when the number of generations is greater than 5000. Assuming that due to special circumstances, the runway capacity of each airport is reduced by 1 / 3 during the 6 o'clock time period, the transit interval of LMN points is 7 minutes, and the transit interval of YIN points is 5 minutes, the final dual-objective results of the coordinated flight release strategy for airport clusters are shown in Table 11.

[0167] Table 11 Results of Dual Optimization Objectives (ω1=0.5, ω2=0.5)

[0168] Total delay time / minutes 15373.44 Total delay cost / yuan 27,535,207 Average delay time in Guangzhou Baiyun District (minutes) 42.70 Average delay time in Bao'an District, Shenzhen (minutes) 36.94 Average delay time in Zhuhai Jinwan (minutes) 7.29 Average delay time in Foshan Shadi (minutes) 9.09 Average delay time in Pingtan, Huizhou / minutes 19.94 Average delay cost in Guangzhou Baiyun District / RMB 77,800 Average delay cost in Bao'an District, Shenzhen (RMB) 64,933 Average delay cost in Zhuhai Jinwan (RMB) 11,867 Average delay cost of Foshan Shadi / Yuan 14,831 Average delay cost in Pingtan, Huizhou / RMB 31,144

[0169] Based on the above dual optimization objectives, it can be concluded that Guangzhou Baiyun Airport has the longest average flight delay time and the highest average flight delay cost; although Zhuhai Jinwan Airport has a considerable number of flights, its average flight delay time and average flight delay cost are at the lowest levels. Therefore, the constraints of the collaborative release model in this application have the least impact on Zhuhai Jinwan Airport, but a greater impact on busy airports such as Guangzhou Baiyun Airport and Shenzhen Bao'an Airport.

[0170] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A method for coordinated flight release and sequencing in an airport cluster, characterized in that: Includes the following steps: S1. The dual optimization objectives are to minimize the total flight delay time and total flight delay cost of the airport cluster, and the constraints are determined to establish a collaborative release model. S2. Obtain flight schedules for all departing airports, preprocess them, and use them as input for the collaborative release model; S3. Use a genetic algorithm to solve for the optimal flight order and realize the coordinated release order of flights in the airport group; The total flight delay time y(t) of the airport group is calculated using the following formula: Where M is the airport group set, m is the airport in the airport group set, and F m For airport m, the collection of departing flights within the next time period T, f im Let t be the i-th departing flight in airport m, and t be the departure time. Let i be the departure time of the i-th departing flight in airport m. Let be the estimated departure time of the i-th departing flight in airport m; Estimated departure time The estimated time to remove wheel chocks is the sum of the airport's average taxiing time, which is the average taxiing time of all departing flights within the selected time period of airport m. The total flight delay cost c(t) of the airport group is calculated using the following formula: Among them, C h The average delay cost per flight for airline h; The total flight delay cost of the airport cluster includes ground delay costs and passenger delay costs, with the ground delay cost per flight per unit time being a. g Calculate using the following formula: Among them, O c The annual operating cost of the airline is N, the number of flights is N, and t1 is the average operating time of each flight, with the average operating time t1 of each flight uniformly set to 2 hours. Passenger delay cost per unit time V p Calculate using the following formula: V p =η·W avg / t2 Where η is the time value coefficient determined based on the differences in travelers' travel purposes, and W avg t2 is the average annual wage of employees nationwide, and t2 is the average effective working hours in one year, t2 = 2000; Passenger delay cost β per unit time for a single flight t Calculate using the following formula: b t =N'·ω·V p Where N' is the number of seats on the flight, and ω is the load factor of the flight.

2. The airport cluster flight coordinated release and sequencing method according to claim 1, characterized in that: The constraints are defined in S1, including: Using runway clearance as a constraint, two consecutively taking off flights must meet the wake turbulence separation standard, expressed by the following formula: t f' ≥t f +s Among them, t f For the departure time of flight f, t f' Let σ be the departure time of flight f' following flight f, and σ be the wake turbulence interval to be maintained.

3. The airport cluster flight coordinated release and sequencing method according to claim 2, characterized in that: The constraints are defined in S1, including: Using runway dynamic hourly capacity as a constraint, the hourly time slots for all flights are obtained. Based on historical flight schedules for peak months, runway departure traffic for each hourly time slot is calculated. For any given hourly time slot, the runway departure traffic for that hourly time slot is ordered, and the runway departure traffic at the 95th quantile is selected as the runway capacity for that hourly time slot. The runway capacity results for all hourly time slots are used as the dynamic hourly runway capacity, satisfying the following formula: p mt' ≤D mt' , Where M is the airport group set, m is the number of airports in the airport group set, and p mt' Let D be the runway departure flow of airport m during time period t'. mt' Let m be the runway capacity of airport m during time period t'.

4. The airport cluster flight coordinated release and sequencing method according to claim 3, characterized in that: The constraints are defined in S1, including: Using the sector's dynamic hourly capacity as a constraint, and based on historical flight schedules for peak months, flights are matched to corresponding hourly time slots for departure traffic aggregation. For any given hourly time slot, the departure traffic within that time slot is ordered, and the departure traffic at the 95th quantile is selected as the departure capacity for that hourly time slot. The sum of traffic flows for all air routes within a sector does not exceed the sector's departure capacity, expressed by the following formula: Where K is the number of routes within the sector, and n j For the traffic flow of each air route in the future time period T, Let S be the off-site capacity of sector S in the future time period T.

5. The airport cluster flight coordinated release and sequencing method according to claim 4, characterized in that: The constraints are defined in S1, including: Using the route through-point interval under flow control as a constraint, the through-point interval of some routes will be restricted due to flow control, that is, a time interval needs to be maintained between adjacent flights, which is expressed by the following formula: in, Let i be the departure time of the i-th and i+1-th departing flights in airport m. These represent the arrival and departure points p of the i-th and (i+1)-th departing flights in airport m, respectively. l Flight time, Ωp l (t”) represents the exit point p. l The release interval during the "t" period.

6. The airport cluster flight coordinated release and sequencing method according to claim 5, characterized in that: The constraints are defined in S1, including: With the constraint that the order of flights passing through the same departure point remains unchanged, the order of flights from different airports must remain consistent, as expressed by the following formula: in, For passing through departure point p l The order of the k-th flight in the original set of flight orders. For passing through departure point p l The order of the k-th flight in the latest set of flight sequences.

7. The airport cluster flight coordinated release and sequencing method according to claim 6, characterized in that: The constraints are defined in S1, including: Using the maximum flight sequence offset as a constraint, to prevent excessive flight adjustments from affecting the normal operation of air traffic control, it is necessary to limit the maximum offset of a single flight sequence, expressed by the following formula: 0≤δ≤20 Where δ is the absolute value of the difference between the flight's index in the original departure sequence set and its index in the latest departure sequence set.

8. The airport cluster flight coordinated release and sequencing method according to claim 1, characterized in that: S3 utilizes a genetic algorithm to obtain the optimal flight order, achieving coordinated flight release ordering for airport clusters, including: S31. Encode the departure delay time series of all departing flights during peak hours in the airport cluster, and treat each feasible departure delay time series as a chromosome, i.e., a population individual. S32. Normalize the dual optimization objectives consisting of total flight delay time and total flight delay cost, and construct a fitness function that meets the release objective by setting a proportional coefficient to reflect the preference of the optimization scheme for reducing delay time and reducing delay cost. S33. Select individuals with large fitness function values ​​from the population as the solution set, and perform crossover recombination on some chromosomes of two parent individuals in the solution set with corresponding probability to form new population individuals. S34. Perform single-point mutation on a departure delay time in the departure delay time series in the solution set with the corresponding probability; S35, proceed to S33, until the termination condition of the genetic algorithm is met, and output the optimal flight order; The fitness function F(t) is expressed by the following formula: Where y(t) is the total flight delay time of the airport group, c(t) is the total flight delay cost of the airport group, and ω1 and ω2 are the ratio coefficients of the total flight delay time and the total flight delay cost, respectively.

Images