A method for quickly identifying strategic conflicts of aircrafts facing airport clusters

Abstract

The present invention discloses a method for quickly identifying strategic conflicts of aircrafts facing an airport group, which includes obtaining historical flight track data and flight plan data, processing the historical flight track data to obtain key kinematic performance parameters of the aircrafts therein; constructing an aircraft kinematic model to calculate all parameters of the four-dimensional coordinates of the aircrafts; constructing a joint distribution model of aircraft kinematic parameters to obtain correlated aircraft performance parameters, which are used as a performance parameter database for deducing aircraft trajectories; obtaining four-dimensional trajectory points of the aircrafts based on the kinematic change rules of approach and departure flight track flight parameters and the navigation coordinate points and time information provided by the flight plan data, thereby completing the deduction of the four-dimensional trajectories of the aircrafts' approach and departure flights; and constructing conflict identification indicators to identify aircraft conflict events. The present invention shortens the time for deducing flight trajectories and identifying conflicts, and realizes the integrated identification of strategic conflicts of aircrafts in an airport group based on flight plan data.

Description

Technical Field

[0001] The present invention belongs to the technical field of civil aviation air traffic control automation and intelligence, and particularly relates to a method for quickly identifying strategic conflicts of aircrafts facing airport clusters. Background Art

[0002] With the increasing of air traffic, the complexity of air traffic management is also constantly increasing. The sharp increase in air traffic flow has led to various problems such as airspace congestion and frequent aircraft conflicts. Traditional conflict detection methods often rely on the analysis of two-dimensional or three-dimensional space and cannot effectively handle complex flight plans and dynamic interactions among multiple airports. Therefore, the present invention uses computer simulation technology to implement a method that can effectively detect flight plan conflicts in four-dimensional space (including the time dimension) to improve flight safety and airspace utilization efficiency.

[0003] Computer simulation technology simulates the actual situation through a scenario model and operation data. When using computer simulation technology to simulate airspace operation and conduct conflict detection, the flight trajectory data of aircrafts based on flight plans is essential. The flight performance database is the performance parameter values of important nodes during the flight of aircrafts and is a necessary database for trajectory deduction. At present, the research on aircraft performance models at home and abroad mainly refers to the statistical data provided by BADA. Although this data has a complete range of aircraft models and is open-source, the actual flight situation of aircrafts is not very consistent with the statistical data in the database, so that the aircraft flight trajectories and conflict allocation deduced on this basis often cannot be perfectly applied to real airports that conform to the national conditions of our country. At the same time, when a large amount of historical flight data is statistically used as the flight performance database, the flight performance parameters are regarded as independently distributed, while there are significant operational interdependencies among aircraft performance parameters. Therefore, the present invention uses aircraft performance parameters based on the joint distribution of a large amount of historical data as the flight performance database.

[0004] Aircraft performance models are widely classified into multiple categories. Among them, the most complex one is the six-degree-of-freedom model, which is usually used for the flight control research of aircrafts. In contrast, the particle model is often used for simplified aircraft flight trajectory research, which ignores roll, pitch, and yaw motions and only considers horizontal and vertical motions. The particle model is divided into two different types: dynamic and kinematic. The main difference between them is that the dynamic model focuses on the effects of forces and energy, while the kinematic model only focuses on the motion state of the aircraft. Generally speaking, the kinematic method is more concise by excluding the influence of forces and is more convenient for calculation. The model proposed by the invention belongs to a pure particle kinematic model, which focuses on the motion state of the aircraft without involving the action of forces, realizes the rapid deduction of the aircraft flight trajectory based on the flight plan, and greatly reduces the time required for trajectory deduction.

[0005] Aircraft strategic conflict identification is a challenging aspect of air traffic flow management. For multi-airport systems (or airport clusters), the aircraft conflict identification process is even more difficult. Due to the close proximity of airport locations, a large number of aircraft, and the serious impact of conflicts, traditional conflict detection methods must be optimized to quickly and accurately locate arrival and departure conflicts. Through the identification of aircraft conflicts in airport clusters under collaborative operation, this invention realizes the integrated monitoring and rapid identification of aircraft conflicts, and further accurately discovers the dangerous proximity between airport clusters, laying a foundation for further eliminating the impact of aircraft flight track conflicts strategically. Summary of the Invention

[0006] Object of the Invention: This invention proposes a method for rapid identification of aircraft strategic conflicts for airport clusters, which shortens the deduction of flight trajectories and the time for conflict identification, and realizes the integrated identification of aircraft strategic conflicts in airport clusters based on flight plan data.

[0007] Technical Solution: A method for rapid identification of aircraft strategic conflicts for airport clusters described in this invention includes the following steps:

[0008] (1) Obtain historical track data and flight plan data, and process the historical track data to obtain the key kinematic performance parameters of the aircraft therein;

[0009] (2) Construct an aircraft kinematic model to obtain all parameters of the aircraft's four-dimensional coordinates;

[0010] (3) Construct a joint distribution model of aircraft kinematic parameters to obtain correlated aircraft performance parameters, serving as a performance parameter database for deducing aircraft trajectories;

[0011] (4) Based on the kinematic change rules of arrival and departure track flight parameters, and based on the navigation coordinate points and time information provided by flight plan data, obtain aircraft four-dimensional track points to complete the deduction of the arrival and departure flight four-dimensional trajectories of the aircraft;

[0012] (5) Construct conflict identification indicators and use a custom heuristic algorithm to identify aircraft conflict events.

[0013] Further, the historical track data in step (1) is divided into three categories: overflight, takeoff, and landing track data; it includes information recording time, flight number, aircraft location information, heading, rate of climb and descent, aircraft speed, aircraft type, and the four-letter codes of the departure and destination airports.

[0014] Further, the implementation process of processing the historical track data in step (1) is as follows:

[0015] Divide arrival and departure data, and consider data with an altitude above 6000 meters as cruise data;

[0016] Delete the abnormal track data with incomplete track points, that is, less than 10 records.

[0017] Separate the track data of different aircraft types according to the aircraft type records in the track data; at the same time, add track numbers to distinguish all tracks; and also consider the flight number and flight date to distinguish the track data of shared flight numbers, add "flight number - flight date" to each track data and use this count as the track number data.

[0018] Furthermore, the flight plan data described in step (1) includes reference points, position information of multiple navigation points, initial flight time information, and aircraft call sign information.

[0019] Furthermore, the implementation process of step (2) is as follows:

[0020] Use a set of ordinary differential equations to construct an aircraft kinematic model:

[0021]

[0022] Among them, V x represents the ground speed of the aircraft, V s represents the vertical airspeed of the aircraft, V R represents the true airspeed of the aircraft, r represents the horizontal movement distance of the aircraft, h represents the vertical movement distance of the aircraft, and t represents the movement time; according to the flight plan data and this model, all parameters (ξ, ψ, η, τ) of the aircraft four-dimensional coordinates can be obtained.

[0023] Furthermore, the implementation process of step (3) is as follows:

[0024] Adopt the Gaussian Copula function as the modeling function for the joint distribution of kinematic parameters; first, use the Gaussian distribution and the gamma distribution to fit V x and V s ; in the two fitted marginal distributions, limit the minimum and maximum values according to historical data, and define the limits of the flight envelope that the aircraft follows during flight.

[0025] For the normal distribution, the probability density function is expressed as:

[0026]

[0027] For the gamma distribution, the probability density function is expressed as:

[0028]

[0029] Among them, μ and σ 2 represent the mean and variance, and α represents the shape of the distribution.

[0030] Secondly, the correlation coefficient matrix among kinematic parameters is obtained; the correlation coefficient matrix of the m-dimensional Gaussian copula function is constructed as follows:

[0031]

[0032] where R represents the m-dimensional correlation coefficient matrix, represents the correlation coefficient of the performance parameter x i with respect to the performance parameter x j , i, j ∈ [1, m]; then a joint distribution model is constructed:

[0033]

[0034] where, is the joint distribution model of the performance parameters x1 to x m , Φ R (q1, q2,..., q m ) is the cumulative distribution function of the performance parameters x1 to x m , and φ R (x1, x2,... x m ) is the m-dimensional normal probability density distribution with mean μ and covariance matrix R, and q i = Φ -1 (u i ); stage random sampling is performed from it to obtain correlated aircraft performance parameters:

[0035] X i = F i -1 (U i ), i = x1, x2,... x m (10)

[0036] The obtained set of is a set of correlated performance parameters; all-stage performance parameter databases are obtained through repeated sampling.

[0037] Furthermore, the implementation process of step (4) is as follows:

[0038] (41) Obtain the navigation point coordinates provided by the flight plan, convert them from latitude and longitude coordinates to coordinates in the Cartesian coordinate system, and determine the heading and key node height of the aircraft according to the navigation points;

[0039] According to the kinematic change law of the approach and departure track flight parameters, obtain the four-dimensional trajectory points of the aircraft; the four-dimensional coordinates of the aircraft are tp(ξ, ψ, η, τ), where ξ represents longitude, ψ represents latitude, η represents altitude, and τ represents time; the four-dimensional trajectory is a set of ordered trajectory points, expressed as:

[0040] P = {tp1, tp2,..., tp n}(11)

[0041] Among them, n represents that this group of trajectories generates n four - dimensional coordinate points. The position information of the reference point and navigation points in the flight plan is transformed into a Cartesian coordinate system with the airport reference point as the origin, and the horizontal distance between each waypoint is calculated; this distance is the distance that the aircraft should fly, and the navigation point coordinates are the coordinates that the aircraft should reach at a specific point:

[0042]

[0043] N Xi , N Yi is the coordinate of a certain navigation point, N X(i+1) , N Y(i+1) is the coordinate of the next navigation point, and d is the horizontal distance between each waypoint;

[0044] (42) Utilize the position information of the navigation points to match the aircraft flight mode into turning and straight - line, and respectively construct a straight - line flight model and a turning flight model;

[0045] Establish a straight - line flight model: Represent the trajectory point coordinates in a coordinate system with the airport reference point as the origin, and obtain the x - coordinate, y - coordinate, and z - coordinate of each trajectory point:

[0046] s i+1 = s i + v i ·sin[arccos(vs i / v i )]·dt·sin(head) (13)

[0047] y i+1 = y i + v i ·sin[arccos(vs i / v i )]·dt·cos(head) (14)

[0048] l = v i ·sin[arccos(vs i / v i )]·dt (15)

[0049] z i+1 = z i + vs i ·dt (16)

[0050] Among them, s i , y i , z i are the coordinates of the current trajectory point; si+1 , y i+1 , z i+1 are the coordinates of the subsequent trajectory point; head is the course between two trajectory points, that is, the direction between the front and rear navigation points, and v i is the speed of the current aircraft, vs i is the climb rate of the current aircraft; dt is the time interval; l is the horizontal distance flown per unit time;

[0051] Establish a turning flight model: When three consecutive waypoints are not on the same straight line, a turning trajectory is required to connect between two straight flight segments; The turning rate of the aircraft uses any turning rate; Determine the turning time according to the course difference between two adjacent flight tracks marked on the aeronautical chart and the turning rate, and the turning range can be obtained by combining the obtained time with the speed; The flight track between every two adjacent track points is approximately regarded as a straight line, and its length is calculated from the speed and the time interval:

[0052] head i+1 = head i ± turning rate · dt (17)

[0053] s i+1 = s i + v i · sin[arccos(vs i / v i )] · dt · sin(head i+1 ) (18)

[0054] y i+1 = y i + v i · sin[arccos(vs i / v i )] · dt · cos(head i+1 ) (19)

[0055]

[0056] z i+1 = z i + vs i · dt (21)

[0057] where head i is the course of the flight segment before the current track point, and head i+1 is the course of the flight segment between the current track point and the next track point. When the aircraft changes its course, a left turn is subtracted and a right turn is added; S is the turning range, and n is the number of track points;

[0058] (43) Smoothly connect the coordinate points to finally complete the four-dimensional trajectory deduction of the aircraft.

[0059] Furthermore, the implementation process of constructing the conflict index in step (5) is as follows:

[0060] When the four-dimensional interval between two trajectories is reduced below the minimum definition standard, a conflict event occurs, and the index is constructed as follows: If there exists a pair of trajectory points tp1(ξ1, ψ1, η1, τ1) ∈ P1 and tp2(ξ2, ψ2, η2, τ2) ∈ P2 that simultaneously satisfy the following minimum vertical interval condition, horizontal interval condition, and time interval condition, then a conflict event occurs:

[0061] |η2 - η1| ≤ Q z (22)

[0062] δ 12 ≤ Q d (23)

[0063] |τ2 - τ1| ≤ Q t (24)

[0064] The distance Q z and Q d are the minimum intervals related to the terminal airspace; the coefficient Q t represents the minimum time interval required to define a conflict event, and δ 12 is the great circle distance between the trajectory points tp1 and tp2.

[0065] Furthermore, the implementation process of identifying aircraft conflict events in step (5) is as follows:

[0066] Prepare the four-dimensional trajectory set F = {P1, P2…P n} of aircraft to be identified. For each trajectory in the sample F, obtain the time of its first trajectory point as a reference for filtering other trajectories in the sample; ensure that each trajectory is only compared with trajectories having similar timestamps. For each trajectory P in F, there exists a subset F i such that where AVG t is the average flight time of the aircraft in the airport group; for each trajectory in the subset F i , compare point by point to satisfy the vertical interval condition, horizontal interval condition, and time interval condition. The conflict events involving the same trajectory among the same points will be discarded; obtain a set of conflict events, and count the conflict events to determine the aircraft conflict points.

[0067] Beneficial effects: Compared with the prior art, the beneficial effects of the present invention are:

[0068] 1. The present invention proposes a combined distribution model for aircraft performance parameters. By comprehensively considering the mutual dependence between parameters, this model constructs an aircraft performance database that better conforms to the actual domestic flight conditions. It can not only simulate the randomness of aircraft performance data during real flights but also effectively improve the accuracy of aircraft performance data, solve the limitations of traditional databases in terms of permissibility and parameter independence, and enhance the accuracy of four-dimensional trajectory kinematic simulations.

[0069] 2. The present invention proposes a method for rapid identification of aircraft conflicts in airport clusters. By customizing a heuristic algorithm to optimize the conflict identification process, it improves the identification speed of arrival and departure aircraft conflicts while ensuring the conflict identification accuracy, provides an effective tool for the safety assessment of airport cluster flight plans, and lays a foundation for further eliminating the impact of aircraft track conflicts strategically. BRIEF DESCRIPTION OF THE DRAWINGS

[0070] Figure 1 is a flowchart of the present invention;

[0071] Figure 2 is a schematic diagram of a turning flight model;

[0072] Figure 3 is a schematic diagram of conflict point identification. DETAILED DESCRIPTION OF THE EMBODIMENTS

[0073] The present invention will be further described below with reference to the drawings.

[0074] As Figure 1 shown, the present invention proposes a method for rapid identification of aircraft strategic conflicts for airport clusters. By establishing a flight performance database based on combined distribution using domestic radar data and performance parameter statistical models, it avoids the strict license agreements attached to existing flight performance databases and conforms to China's flight standards; uses a kinematic model to deduce the flight trajectory based on the flight plan, and uses a heuristic algorithm to quickly detect aircraft conflict events under the flight plan, shortening the deduction of the flight trajectory and the identification time of conflict identification, and realizing an integrated method for identifying aircraft strategic conflicts in airport clusters based on flight plan data. The specific steps are as follows:

[0075] Step 1: Obtain historical track data and flight plan data, and process the historical track data to obtain the key kinematic performance parameters of the aircraft.

[0076] Historical track data includes overflight, takeoff, and landing tracks. Extract the takeoff and landing track data of all airports belonging to a specific airport group, that is, all tracks taking off and landing at an airport within a region centered on the airport with a radius of 200 kilometers, including information recording time, flight number, aircraft position information (longitude, latitude, and altitude), heading, rate of climb and descent, aircraft speed, aircraft type, etc. And divide the arrival and departure data, where data with an altitude above 6000 meters is regarded as cruise data. All subsequent steps are processed separately for arrival and departure trajectories.

[0077] Considering the instability of the equipment, delete the track data with incomplete track points (less than 10 records), and this track is considered an abnormal track caused by equipment problems.

[0078] According to the aircraft type records in the track data, distinguish the track data of different aircraft types. At the same time, add a track number to distinguish all tracks. There may be multiple track data corresponding to the same flight number, so the flight number alone cannot be used as the distinguishing label for the tracks. However, to prevent air traffic control chaos, this situation rarely occurs on the same day. Therefore, considering both the flight number and the flight date can distinguish such track data with shared flight numbers. Add "flight number - flight date" to each track data and use this as the track number data for counting, so that all track data can be distinguished from each other.

[0079] Prepare the flight plan data for the corresponding airport (airport group) as follows: including reference points, position information (longitude, latitude, and altitude) of multiple navigation points. The navigation points are sorted in order, and generally, the number of navigation point information is between 5 and 8; initial flight time information, where the departure flight plan is the takeoff time, and the arrival flight plan is the time of navigation point 1. The time information includes year, month, day, hour, minute, and second; and aircraft call sign information.

[0080] Step 2: Build an aircraft kinematic model to obtain all parameters of the aircraft's four-dimensional coordinates.

[0081] Regardless of the flight phase of the aircraft (arrival, departure, and cruise phases), the aircraft flight trajectory can be constructed using a set of ordinary differential equations:

[0082]

[0083] Among them, V x represents the ground speed of the aircraft, V s represents the vertical airspeed of the aircraft, V R represents the true airspeed of the aircraft, r represents the horizontal movement distance of the aircraft, h represents the vertical movement distance of the aircraft, and t represents the movement time. Let V x and V sAre regarded as the kinematic parameters of the aircraft. This model is the basic model for constructing the four-dimensional coordinates of the aircraft. According to the flight plan data and using this model, all the parameters (ξ, ψ, η, τ) of the four-dimensional coordinates of the aircraft can be obtained.

[0084] Step 3: Construct a joint distribution model of the kinematic parameters of the aircraft to obtain the correlated aircraft performance parameters, which are used as the performance parameter database for deducing the aircraft trajectory.

[0085] The Gaussian Copula function is used as the modeling function for the joint distribution of kinematic parameters. First, the Gaussian distribution and the gamma distribution are used to fit V x and V s . In the two fitted marginal distributions, the minimum and maximum values are restricted according to historical data, and the limits of the flight envelope that the aircraft follows during flight are defined. For the normal distribution, the probability density function is expressed as:

[0086]

[0087] For the gamma distribution, the probability density function is expressed as:

[0088]

[0089] where μ and σ 2 represent the mean and variance, and α represents the shape of the distribution.

[0090] Secondly, the correlation coefficient matrix between kinematic parameters is obtained. The correlation coefficient matrix of the m-dimensional Gaussian copula function is constructed as follows:

[0091]

[0092] where R represents the m-dimensional correlation coefficient covariance matrix, represents the correlation coefficient of the performance parameter x i relative to the performance parameter x j , i, j ∈ [1, m]. Then the joint distribution model can be constructed:

[0093]

[0094] V ∼ N(μ, R) (9)

[0095] where, is the joint distribution model of the performance parameters x1 to x m , Φ R (q1, q2,..., q m ) is the cumulative distribution function of the performance parameters x1 to x m , and φ R (x1, x2,... x m) is an m - dimensional normal probability density distribution with mean μ and covariance matrix R, and q i = Φ -1 (u i ). From this, a joint distribution model of aircraft kinematic parameters is obtained. Sampling is carried out 5000 times for each stage, and correlated aircraft performance parameters are obtained as the performance parameter database for generating aircraft trajectories. During the simulation, random sampling is performed on the performance parameters to simulate the randomness of different flight speeds of different flights during the actual flight process. The sampling process is as follows:

[0096] X i = F i -1 (U i ), i = x1, x2,... x m (10)

[0097] The obtained set of is a set of correlated performance parameters. By repeating the sampling, the performance parameter database for all stages can be obtained.

[0098] Step 4: According to the kinematic change law of the approach and departure track flight parameters, based on the navigation coordinate points and time information provided by the flight plan data, obtain the four - dimensional trajectory points of the aircraft and complete the four - dimensional trajectory deduction of the aircraft's approach and departure flight

[0099] First, obtain the navigation point coordinates provided by the flight plan, and make the aircraft fly according to the longitude and latitude coordinates of the navigation points of the actual approach and departure flight procedures. On the basis of establishing the aircraft performance database based on joint distribution sampling, according to the kinematic change law of the approach and departure track flight parameters, obtain the four - dimensional trajectory points of the aircraft, and finally complete the four - dimensional trajectory deduction of the aircraft's approach and departure flight.

[0100] Use the flight trajectory model and different flight modes to calculate the aircraft coordinate points, arrange them in order, and obtain the four - dimensional coordinates of the aircraft as tp(ξ, ψ, η, τ), where ξ represents longitude, ψ represents latitude, η represents altitude, and τ represents time. The four - dimensional trajectory is a set of ordered trajectory points, where n represents that this trajectory generates n four - dimensional coordinate points, expressed as:

[0101] P = {tp1, tp2,..., tp n}} (11)

[0102] Convert the position information (longitude, latitude, and altitude) of the reference point and navigation points in the flight plan to the Cartesian coordinate system with the airport reference point as the origin, and calculate the horizontal distance between each section of waypoints. This distance is the distance that the aircraft should fly, and the navigation point coordinates are the coordinates that the aircraft should reach at a specific point.

[0103]

[0104] Among them, N Xi , N Yi is the coordinate of a certain navigation point, N X(i+1) , N Y(i+1) is the coordinate of the next navigation point, and d is the horizontal distance between each waypoint.

[0105] Match the flight mode, and use the position information of the navigation points to match the flight mode of the aircraft to turning and straight flight, and establish a straight flight model and a turning flight model respectively.

[0106] Establish a straight flight model: Represent the trajectory point coordinates in a coordinate system with the airport reference point as the origin, and obtain the x coordinate, y coordinate, and z coordinate of each trajectory point:

[0107] s i+1 = s i + v i ·sin[arccos(vs i / v i )]·dt·sin(head) (13)

[0108] y i+1 = y i + v i ·sin[arccos(vs i / v i )]·dt·cos(head) (14)

[0109] l = v i ·sin[arccos(vs i / v i )]·dt (15)

[0110] z i+1 = z i + vs i ·dt (16)

[0111] Among them, s i , y i , z i are the coordinates of the current trajectory point; s i+1 , y i+1 , z i+1 are the coordinates of the next trajectory point; head is the course between the two trajectory points, that is, the direction between the front and rear navigation points, v i is the current aircraft speed; dt is the time interval; l is the horizontal distance flown per unit time.

[0112] Establish as Figure 2The turning flight model shown: When three consecutive waypoints are not on the same straight line, a turning trajectory is required to connect between two straight flight segments. In this embodiment, the turning rate of the aircraft uses the standard turning rate of 3 degrees per second. The turning time is determined based on the course difference between two adjacent flight tracks marked on the aeronautical chart and the turning rate. The turning range can be obtained by combining the obtained time with the speed. The flight track between every two adjacent waypoints is approximately regarded as a straight line, and its length is calculated from the speed and the time interval.

[0113] head i+1 = head i ± turning rate·dt (17)

[0114] s i+1 = s i + v i ·sin[arccos(vs i / v i )]·dt·sin(head iv1 ) (18)

[0115] y i+1 = y i + v i ·sin[arccos(vs i / v i )]·dt·cos(head i+1 ) (19)

[0116]

[0117] z i+1 = z i + vs i ·dt (21)

[0118] Among them, head i is the course of the flight segment before the current waypoint, head i+1 is the course of the flight segment between the current waypoint and the next waypoint. When the aircraft changes its course, a left turn is subtraction and a right turn is addition; S is the turning range, and n is the number of waypoints. Smoothly connect the coordinate points to finally complete the four-dimensional flight track deduction of the aircraft.

[0119] Step 5: Construct conflict recognition indicators and use a custom heuristic algorithm to quickly identify aircraft conflict events under the flight plan deduction track.

[0120] A conflict event occurs when the four - dimensional interval between two trajectories drops below the minimum definition standard. The index is constructed as follows: If there exists a pair of trajectory points tp1(ξ1, ψ1, η1, τ1) ∈ P1 and tp2(ξ2, ψ2, η2, τ2) ∈ P2 that simultaneously satisfy the following minimum vertical - separation condition, horizontal - separation condition, and time - separation condition, then a conflict event occurs.

[0121] |η2 - η1| ≤ Q z Vertical - separation condition (C z ) (22)

[0122] δ 12 ≤ Q d Horizontal - separation condition (C d ) (23)

[0123] |τ2 - τ1| ≤ Q t Time - separation condition (C t ) (24)

[0124] The distance Q z and Q d are the minimum separation values related to the terminal airspace. The coefficient Q t represents the minimum time interval required to define a conflict event, and δ 12 is the great - circle distance between the trajectory points tp1 and tp2.

[0125] Considering a set of trajectories F = {P1, P2…P n} as a flight - route sample, the conflict - event recognition method consists of four steps:

[0126] 1) Pre - processing: Prepare the four - dimensional trajectory set F = {P1, P2…P n} of aircraft to be recognized, where tp i (ξ ι , ψ ι , η ι , τ ι ) ∈ P i , i ∈ [1, n].

[0127] 2) Filtering: For each trajectory in the sample F, obtain the time of its first trajectory point as a reference for filtering the other trajectories in the sample. This process ensures that each trajectory is only compared with trajectories having similar timestamps. Thus, for each trajectory P in F, there exists a subset F i such that where AVG t is the average flight time of the aircraft in the airport group.

[0128] 3) Searching: For each trajectory in the subset F i , compare point - by - point to satisfy the separation condition Cd , C z and C t . The conflict events involving the same trajectory among the similarities will be discarded.

[0129] 4) Result: Obtain a set of conflict events, and count the conflict events to determine the conflict points of the aircraft.

[0130] Repeat steps 2 and 3 for each trajectory interdependent area identified in the airport group. Finally, the specific locations and times of aircraft conflicts can be determined, as Figure 3 shown. For the track data generated from the arrival and departure flight plans between airport groups, conflict identification is performed. For aircraft that do not meet the separation conditions in both the horizontal and vertical directions, the time interval also needs to be judged. The points that meet the conflict conditions will be recorded (black dots, black triangles), and the points that meet the spatial separation but do not meet the time conditions (i.e., trajectory conflicts but no conflicts in time) will not be recorded (gray dots).

[0131] Select the airport group in the Guangdong-Hong Kong-Macao Greater Bay Area for case verification. Select the flight plan deduction trajectories of five airports: Guangzhou Baiyun Airport (ZGGG), Shenzhen Bao'an Airport (ZGSZ), Zhuhai Jinwan Airport (ZGSD), Hong Kong International Airport (VHHH), and Macau International Airport (VMMC), and use the present invention to check the conflict points between the trajectories.

[0132] Select the flight plan data of the airport group in the Guangdong-Hong Kong-Macao Greater Bay Area on May 2, 2023, to generate a set of four-dimensional trajectories of aircraft P = {P1, P2... P n}, n = 2171, that is, there are a total of 2171 flight plan deduction tracks. Among them, there are 1114 departure flights and 1057 arrival flights. The time span is from 00:00 on May 20, 2023, to 00:35 on May 21, 2023. The average flight time AVG t set within the airport group is 15 minutes. Finally, a total of 396 flight plans with conflicts are detected, and several specific conflict points.

Claims

1. A rapid identification method for strategic conflicts of aircrafts facing airport clusters, characterized in that, It includes the following steps: (1) Obtain historical track data and flight plan data, and process the historical track data to obtain the key kinematic performance parameters of the aircraft therein; (2) Construct an aircraft kinematic model to obtain all parameters of the four-dimensional coordinates of the aircraft; (3) Construct a joint distribution model of aircraft kinematic parameters to obtain correlated aircraft performance parameters, serving as a performance parameter database for deducing the aircraft trajectory; (4) According to the kinematic change law of the approach and departure track flight parameters, based on the navigation coordinate points and time information provided by the flight plan data, obtain the four-dimensional track points of the aircraft, and complete the four-dimensional track deduction of the aircraft's approach and departure flight; (5) Construct conflict recognition indicators and use a custom heuristic algorithm to identify aircraft conflict events; The implementation process of step (3) is as follows: The Gaussian Copula function is used as the modeling function for the joint distribution of kinematic parameters; first, the Gaussian distribution and the gamma distribution are used to fit the ground speed V of the aircraft x and the vertical airspeed V of the aircraft s ; in the two fitted marginal distributions, the minimum and maximum values are restricted according to historical data, and the limits of the flight envelope that the aircraft must comply with during flight are defined; For the normal distribution, the probability density function is expressed as: For the gamma distribution, the probability density function is expressed as: where μ and σ 2 represent the mean and variance, and α represents the shape of the distribution; Secondly, obtain the correlation coefficient matrix between kinematic parameters; the correlation coefficient matrix of the m-dimensional Gaussian copula function is constructed as follows: where R represents an m-dimensional correlation coefficient matrix, represents the performance parameter x i with respect to the performance parameter x j correlation coefficient, i, j ∈ [1, m]; then a joint distribution model is constructed: V~N(μ,R) (9) Among them, is the joint distribution model of performance parameters x1 to x m , Φ R (q1, q2,..., q m ) is the cumulative distribution function of performance parameters x1 to x m , while φ R (x1, x2,... x m ) is an m-dimensional normal probability density distribution with a mean of μ and a covariance matrix of R, and q i = Φ -1 (u i ); stage random sampling is performed from it to obtain correlated aircraft performance parameters: The obtained set is a set of correlated performance parameters; the performance parameter database for all stages is obtained through repeated sampling.

2. The rapid identification method for aircraft strategic conflicts facing an airport group according to claim 1, characterized in that The historical track data in step (1) is divided into three categories: overflight, takeoff, and landing track data; it includes information recording time, flight number, aircraft location information, heading, rate of climb and descent, aircraft speed, aircraft type, and the four-character codes of the departure and destination airports.

3. The rapid identification method for aircraft strategic conflicts facing airport clusters according to claim 1, characterized in that The implementation process of processing the historical track data in step (1) is as follows: Divide the approach and departure data, and consider the data with a height above 6000 meters as cruise data; Delete abnormal track data with incomplete track points, that is, less than 10 records; According to the aircraft type records in the track data, distinguish the track data of different aircraft types; at the same time, add track numbers to distinguish all tracks; and also consider the flight number and flight date to distinguish the track data of shared flight numbers, add "flight number - flight date" to each track data and use this as the track number data for counting.

4. A method for quickly identifying strategic conflicts of aircrafts for airport groups according to claim 1, characterized in that, The flight plan data in step (1) includes reference points, location information of multiple navigation points, initial flight time information, and aircraft call sign information.

5. The rapid identification method for aircraft strategic conflicts facing airport clusters according to claim 1, characterized in that The implementation process of step (2) is as follows: Use a set of ordinary differential equations to construct an aircraft kinematic model: Among them, V x represents the ground speed of the aircraft, V s represents the vertical airspeed of the aircraft, V R represents the true airspeed of the aircraft, r represents the horizontal movement distance of the aircraft, h represents the vertical movement distance of the aircraft, and t represents the movement time; all parameters (ξ, ψ, η, τ) of the four-dimensional coordinates of the aircraft can be obtained according to the flight plan data and this model, where ξ represents longitude, ψ represents latitude, η represents altitude, and τ represents time.

6. The method for quickly identifying strategic conflicts of aircrafts facing an airport group according to claim 1, characterized in that The implementation process of step (4) is as follows: (41) Obtain the navigation point coordinates provided by the flight plan, convert them from longitude and latitude coordinates to coordinates in the Cartesian coordinate system, and determine the heading and key node height of the aircraft according to the navigation points; According to the kinematic change law of the approach and departure track flight parameters, obtain the four-dimensional track points of the aircraft; the four-dimensional coordinates of the aircraft are tp(ξ, ψ, η, τ), where ξ represents longitude, ψ represents latitude, η represents altitude, and τ represents time; the four-dimensional track is a set of ordered track points, expressed as: P = {tp1, tp2,..., tp n}(11) Among them, n represents that this group of trajectories generates n four-dimensional coordinate points. The position information of the reference point and navigation points in the flight plan is transformed into a Cartesian coordinate system with the airport reference point as the origin, and the horizontal distance between each waypoint is calculated; this distance is the distance that the aircraft should fly, and the navigation point coordinates are the coordinates that the aircraft should reach at a specific point: N Xi ,N Yi is the coordinate of a certain navigation point, N X(i+1) ,N Y(i+1) is the coordinate of the next navigation point, and d is the horizontal distance between each waypoint; (42) Use the position information of the navigation points to match the aircraft flight mode as turning and straight, and construct a straight flight model and a turning flight model respectively; Establish a straight flight model: Represent the track point coordinates in a coordinate system with the airport reference point as the origin, and obtain the x coordinate, y coordinate, and z coordinate of each track point: s i+1 = s i + v i · sin[arccos(vs i / v i )]· dt· sin(head) (13) y i+1 = y i + v i · sin[arccos(vs i / v i )]· dt· cos(head) (14) l = v i ·sin[arccos(vs i / v i )]·dt (15) z i+1 = z i + vs i · dt (16) Among them, s i , y i , z i are the coordinates of the current trajectory point; s i+1 , y i+1 , z i+1 are the coordinates of the next trajectory point; head is the course between the two trajectory points, that is, the direction between the front and rear navigation points, and v i is the current aircraft speed, and vs i is the current aircraft climb rate; dt is the time interval; l is the horizontal distance flown per unit time; Establish a turning flight model: When three consecutive waypoints are not on the same straight line, a turning track is required to connect two straight segments; the aircraft's turning rate uses an arbitrary turning rate; the turning time is determined based on the heading difference and turning rate between two adjacent segments of the track marked on the chart, and the turning distance can be obtained by combining the obtained time with the speed; the track between every two adjacent track points is approximately regarded as a straight line, and its length is calculated from the speed and time interval: head i+1 = head i ± turning rate · dt (17) s i+1 = s i + v i ·sin[arccos(vs i / v i )]·dt·sin(head i+1 ) (18) y i+1 = y i + v i · sin[arccos(vs i / v i )]· dt· cos(head i+1 ) (19) z i+1 = z i + vs i ·dt (21) Among them, head i is the course of the previous leg before the current waypoint, and head i+1 is the course of the leg between the current waypoint and the next waypoint. When the aircraft changes its course, a left turn is subtracted and a right turn is added; S is the turning distance, and n is the number of waypoints; (43) Smoothly connect the coordinate points and finally complete the four-dimensional trajectory deduction of the aircraft.

7. A method for quickly identifying strategic conflicts of aircrafts facing an airport group according to claim 1, characterized in that, The implementation process of constructing the conflict indicator in step (5) is as follows: A conflict event occurs when the four-dimensional separation of two trajectories drops below the minimum definition standard. The indicator is constructed as follows: If there is a pair of trajectory points tp1(ξ1, ψ1, η1, τ1)∈P1 and tp2(ξ2, ψ2, η2, τ2)∈P2 that simultaneously satisfy the following minimum vertical separation conditions, horizontal separation conditions, and time interval conditions, a conflict event occurs: |η2 - η1| ≤ Q z (22) δ 12 ≤Q d (23) |τ2 - τ1| ≤ Q t (24) Distance Q z and Q d is the minimum separation related to the terminal airspace; Coefficient Q t represents the minimum time interval δ required to define a conflict event 12 is the great circle distance between the trajectory points tp1 and tp2.

8. The rapid identification method for aircraft strategic conflicts facing an airport group according to claim 1, characterized in that, The process of identifying the aircraft conflict event in step (5) is as follows: Prepare the four-dimensional trajectory set F of the aircraft to be recognized, F = {P1, P2... P n}, for each trajectory in the sample F, obtain the time of its first trajectory point as a reference for filtering other trajectories in the sample; ensure that each trajectory is only compared with trajectories having similar timestamps. For each trajectory P in F, there exists a subset F i of the trajectory, such that where AVG t is the average flight time of the aircraft among the airport group; for each trajectory in the subset F i , compare point by point to meet the vertical separation condition, horizontal separation condition and time separation condition, and conflict events involving the same trajectory among the same points will be discarded; A set of conflict events is obtained, and the conflict events are counted to determine the aircraft conflict points.

Images