Direct flight automatic identification method based on QAR data

Abstract

The invention discloses a direct flight automatic identification method based on QAR data, and the method comprises the steps: building a t-SNE model confusion degree through the human-computer interaction information of different longitudes and latitudes and altitude flight path deviation distances in a flight stage, and building a flight stage flight path deviation distance automatic identification algorithm through the t-SNE model confusion degree. Therefore, the position coordinate man-machine interaction information of the flight stage is automatically identified by using a flight stage flight path deviation distance automatic identification algorithm, and the safety limit interval data within the flight path deviation distance calibration time of the flight stage in an automatic driving state is obtained. And setting course, flight speed and flight attitude correction schemes according to the attitude stability judgment information, and performing aircraft safety control based on the course, flight speed and flight attitude correction schemes. According to the method, the automatic identification precision of automatic identification of the flight path deviation distance in the flight stage is improved. According to the method, a large amount of flight data can be efficiently processed, the intelligent level of flight management is improved, and the operation complexity is reduced.

Description

Technical Field

[0001] The invention relates to the field of automatic identification of direct flights of aircraft, and in particular to a method for automatic identification of direct flights based on QAR data. Background Art

[0002] Direct flight, in the process of flight, means that the aircraft is allowed to fly directly between waypoints under the premise of complying with route rules, due to control commands, open airspace, good weather conditions or other factors, skipping some waypoints in the original plan, thereby shortening the flight distance and time. According to the actual situation of airlines in fuel efficiency evaluation, direct flight has attracted much attention because it can intuitively save actual flight distance, thereby most significantly saving fuel consumption and achieving the purpose of fuel saving. Therefore, how to automatically identify and discover direct flights has become a key problem that troubles airlines.

[0003] There are two main methods for analyzing direct flights in the traditional civil aviation field. One is based on manual visual inspection, that is, by observing the actual flight trajectory of the flight, comparing it with the flight plan trajectory, and finding that there is a deviation from the track, further observing whether there is an obvious distance saving to determine whether the flight is a direct flight. At the same time, a flight may have multiple such deviations, and careful judgment and screening are required to determine the actual direct flights and flight segments. The accuracy of this method is greatly affected by human factors, and it requires considerable experience to accurately draw conclusions.

[0004] Another method is based on inquiry, asking pilots whether they have applied for direct flights at certain locations or judging at which locations the pilots have applied for direct flights through cockpit voice, and then contacting the Air Traffic Control Bureau and other relevant departments through instant messaging, telephone, telegram, etc. to verify whether direct flights have actually been made at that location. This method is time-consuming and labor-intensive, and is prone to errors caused by memory bias, and is also difficult to verify. Summary of the invention

[0005] The present invention overcomes the shortcomings of the prior art and provides a direct flight automatic identification method based on QAR data.

[0006] To achieve the above object, the technical solution adopted by the present invention is a direct flight automatic identification method based on QAR data, comprising the following steps:

[0007] Establishing a QAR data integration center, and using the QAR data integration center to set flight trajectory parameters, obtaining the QAR data integration center after the flight trajectory parameters are set, and using the QAR data integration center to control the position coordinate human-computer interaction information of the QAR integrated flight phase; the flight phase includes a take-off phase, a climb phase, a cruise phase, and a descent phase;

[0008] The direct flight information database is used to integrate the human-computer interaction information of the track deviation distance at different latitudes, longitudes and altitudes during the flight phase, and the t-SNE model perplexity is established through the human-computer interaction information of the track deviation distance at different latitudes, longitudes and altitudes during the flight phase. The automatic identification algorithm of the track deviation distance during the flight phase is established through the t-SNE model perplexity;

[0009] Automatically identify the human-computer interaction information of the position coordinates during the flight phase using the automatic identification algorithm for the track deviation distance during the flight phase, obtain the safety limit interval data within the calibration time of the track deviation distance during the flight phase under the automatic driving state, and judge the attitude stability through the safety limit interval data within the calibration time of the track deviation distance during the flight phase under the automatic driving state, and generate attitude stability judgment information;

[0010] The correction plan for heading, flight speed and flight attitude is set through attitude stability judgment information, and the aircraft is safely controlled based on the correction plan for heading, flight speed and flight attitude.

[0011] Furthermore, in the present method, a QAR data integration center is established, and flight trajectory parameters are set for the QAR data integration center to obtain the QAR data integration center after the flight trajectory parameters are set, specifically including:

[0012] Establish a QAR data integration center, and test the QAR data integration center to obtain the ADS-B direct flight parameter jump interval characteristics at different latitudes, longitudes and altitudes of the QAR data integration center, and establish an ADS-B direct flight parameter boundary threshold evaluation model based on the track monitoring system;

[0013] Input the ADS-B direct flight parameter jump interval characteristics of different latitudes, longitudes and altitudes of the QAR data integration center into the ADS-B direct flight parameter boundary threshold evaluation model for training, and obtain the trained ADS-B direct flight parameter boundary threshold evaluation model;

[0014] Obtain the ADS-B direct flight parameter jump interval characteristics of the QAR data integration center within the preset time and input them into the trained ADS-B direct flight parameter boundary threshold evaluation model for evaluation, and obtain the ADS-B direct flight parameter boundary threshold of the QAR data integration center at a unit timestamp;

[0015] The real-time ADS-B direct flight parameters of the QAR data integration center are obtained. When the real-time ADS-B direct flight parameters are greater than the ADS-B direct flight parameter boundary threshold of the QAR data integration center at the unit timestamp, the real-time ADS-B direct flight parameters are adjusted according to the ADS-B direct flight parameter boundary threshold of the QAR data integration center at the unit timestamp to obtain the QAR data integration center after the flight trajectory parameters are set.

[0016] Furthermore, in this method, the direct flight information database is used to integrate the human-computer interaction information of the track deviation distance at different latitudes and longitudes and altitudes in the flight phase, and the t-SNE model perplexity is established through the human-computer interaction information of the track deviation distance at different latitudes and longitudes and altitudes in the flight phase. The automatic recognition algorithm of the track deviation distance in the flight phase is established through the t-SNE model perplexity, which specifically includes:

[0017] The direct flight information database is used to integrate the human-computer interaction information of the track deviation distance at different latitudes, longitudes and altitudes during the flight phase, and the track deviation distance type is classified using the human-computer interaction information of the track deviation distance at different latitudes, longitudes and altitudes during the flight phase, and the human-computer interaction information of each track deviation distance type is obtained. The t-SNE model perplexity of each track deviation distance type is established through the human-computer interaction information of each track deviation distance type;

[0018] The human-computer interaction information in the t-SNE model perplexity of each track deviation distance type is input into the safety limit interval clustering calculation model, the safety limit interval within the track deviation distance calibration time existing in the human-computer interaction information per unit time is obtained, and it is determined whether the safety limit interval within the track deviation distance calibration time existing in the human-computer interaction information is equal to or exceeds the preset interval, and the DE-9IM model is introduced;

[0019] When the safety limit interval within the track deviation distance calibration time existing in the human-computer interaction information is equal to or exceeds the preset interval, the safety limit interval within the track deviation distance calibration time corresponding to the safety limit interval information of the track deviation distance type of the unit time human-computer interaction information is obtained as the relevant safety limit interval within the track deviation distance calibration time; when the safety limit interval within the track deviation distance calibration time existing in the human-computer interaction information is less than the preset interval, the safety limit interval within the track deviation distance calibration time corresponding to the safety limit interval information of the track deviation distance type of the unit time human-computer interaction information is used as the relevant safety limit interval within the track deviation distance calibration time;

[0020] The relevant track deviation distance calibration time safety limit interval is input into the DE-9IM model, so that the input dimension is concentrated in the relevant track deviation distance calibration time safety limit interval, the preset position node is obtained, and the flight phase track deviation distance automatic identification algorithm is established based on the route curvature, and the preset position node in the t-SNE model perplexity is input into the flight phase track deviation distance automatic identification algorithm for deviation warning;

[0021] The waypoints given by the computer flight plan trajectory are projected to the map coordinate system, and circular buffers are drawn for all waypoints according to the maximum offset distance of the route given by the Civil Aviation Administration. The actual flight trajectory of the QAR is projected to the map coordinate system, and the lines connecting the longitude and latitude trajectory points are formed. The DE-9IM performs logical operations to determine whether the spatial relationship between the actual flight trajectory and the circular buffer of the waypoints intersects, and iterates through the circular buffers of all waypoints to determine whether some waypoints are not passed, and the number of consecutive points not passed is greater than or equal to 1, it is considered that the flight trajectory deviates;

[0022] Extract the starting waypoint and ending waypoint of the flight plan from which the actual flight trajectory deviates, and find the nearest QAR actual flight trajectory point through Euclidean distance to intercept the actual flight trajectory;

[0023] By using the route curvature calculation method, the flight distance of the actual flight trajectory is compared with the straight-line distance from the starting waypoint to the ending waypoint. If the curvature percentage is less than a given threshold, it is determined whether it is a direct flight. The threshold can be set by yourself, and the saved distance is obtained by subtracting the flight distance of the actual flight trajectory from the flight distance of the computer flight plan; Route curvature (%) = (flight distance of actual flight trajectory /

[0024] The straight-line distance from the starting waypoint to the ending waypoint) × 100%.

[0025] Furthermore, in this method, the position coordinate human-computer interaction information of the flight phase is automatically identified by using the automatic identification algorithm of the flight phase track deviation distance, and the safety limit interval data within the calibration time of the track deviation distance of the flight phase in the automatic driving state is obtained. The attitude stability is judged by the safety limit interval data within the calibration time of the track deviation distance of the flight phase in the automatic driving state, and the attitude stability judgment information is generated, which specifically includes:

[0026] The human-computer interaction information of the position coordinates of the flight phase is input into the automatic identification algorithm of the track deviation distance of the flight phase for automatic identification, and the data of the safe limited interval within the calibration time of the track deviation distance of the flight phase in the automatic driving state is obtained, and the attitude stability judgment deflection angle standard is preset;

[0027] The attitude stability judgment declination standard is used to divide the attitude stability judgment declination data of the safe limited interval within the track deviation distance calibration time during the flight phase under the automatic driving state, obtain the attitude stability judgment declination of the safe limited interval within the track deviation distance calibration time during the flight phase under the automatic driving state, and integrate the navigation deviation information with different magnetic declination angles during the flight phase;

[0028] The attitude stability judgment is generated by judging the declination of the attitude stability within the safe limited interval within the calibration time of the track deviation distance during the flight phase in the automatic driving state, the navigation deviation information with different magnetic declination angles during the flight phase, and the data of the safe limited interval within the calibration time of the track deviation distance during the flight phase in the automatic driving state, and the attitude stability judgment information is generated.

[0029] Furthermore, in this method, the correction scheme of heading, flight speed and flight attitude is set by judging the attitude stability, which specifically includes:

[0030] The navigation deviation information of different magnetic declination angles in the flight phase is integrated through attitude stability judgment information, and the fixed magnetic declination influence position is established based on the navigation deviation information of different magnetic declination angles in the flight phase. The fixed point correction is performed through the fixed magnetic declination influence position to integrate the track deviation of the flight phase;

[0031] The flight control system error is obtained through the track deviation during the flight phase, and the flight control system error risk coefficient is established. The flight control system error is input into the flight control system error risk coefficient for visualization to obtain risk assessment results of different levels;

[0032] Based on the risk assessment results of different levels, correction plans for heading, flight speed and flight attitude during direct flight under different magnetic declination angles are set, and the correction plans for heading, flight speed and flight attitude are output.

[0033] Furthermore, in this method, the aircraft safety control is performed based on the correction scheme of heading, flight speed and flight attitude, specifically including:

[0034] Obtain the distribution characteristic information of turbulence frequency of different airflow types per unit time and the navigation deviation information of different magnetic declination angles in a specific area, and calculate the Cronbach's alpha coefficient between the areas where the turbulence frequency of different airflow types is located and the navigation deviation information of different magnetic declination angles in the flight stage through the distribution characteristic information of turbulence frequency of different airflow types per unit time and the navigation deviation information of different magnetic declination angles in the flight stage;

[0035] Generative adversarial networks were introduced, and weight matrices were set through the generative adversarial networks. The turbulence frequency data of different airflow types were initialized through the Cronbach's α coefficient between the turbulence frequency areas of different airflow types and the navigation deviation information with different magnetic declinations in the flight phase, and the dangerous calibration areas of the turbulence frequency of each different airflow type were obtained.

[0036] Determine whether the Cronbach's alpha coefficient between the areas where the turbulence frequencies of different airflow types are located and the navigation deviation information with different magnetic declinations in the flight phase is greater than a preset Cronbach's alpha coefficient; when the Cronbach's alpha coefficient between the areas where the turbulence frequencies of different airflow types are located and the navigation deviation information with different magnetic declinations in the flight phase is not greater than the preset Cronbach's alpha coefficient, output the danger calibration area of ​​each different airflow type turbulence frequency;

[0037] When the Cronbach's alpha coefficient between the areas where the turbulence frequencies of different airflow types are located and the navigation deviation information with different magnetic declinations in the flight stages is greater than the preset Cronbach's alpha coefficient, the dangerous calibration areas with different turbulence frequencies of different airflow types will be screened in a focused manner until the Cronbach's alpha coefficient between the areas where the turbulence frequencies of different airflow types are located and the navigation deviation information with different magnetic declinations in the flight stages is no greater than the preset Cronbach's alpha coefficient.

[0038] Beneficial effects:

[0039] The present invention uses a QAR data integration center to establish and set flight trajectory parameters for the QAR data integration center, obtains the QAR data integration center after the flight trajectory parameters are set, uses the QAR data integration center to control the QAR to integrate the human-computer interaction information of the position coordinates of the flight stage, and then uses a direct flight information database to integrate the human-computer interaction information of different longitudes and latitudes and altitudes of the flight stage, and establishes a t-SNE model perplexity through the human-computer interaction information of the track deviation distance at different longitudes and latitudes and altitudes of the flight stage, and establishes a flight stage track deviation distance automatic recognition algorithm through the t-SNE model perplexity, so as to automatically recognize the human-computer interaction information of the position coordinates of the flight stage through the flight stage track deviation distance automatic recognition algorithm, obtain the track deviation distance safety limit interval data of the flight stage in the automatic driving state within the calibration time, and perform attitude stability judgment through the track deviation distance safety limit interval data of the flight stage in the automatic driving state within the calibration time, generate attitude stability judgment information, and finally set a correction scheme of heading, flight speed and flight attitude through the attitude stability judgment information, and perform aircraft safety control based on the correction scheme of heading, flight speed and flight attitude. The present invention uses the DE-9IM model to process the target track deviation distance in the t-SNE model perplexity within the safe limited interval within the calibration time, so that the input dimension is concentrated in the target track deviation distance in the t-SNE model perplexity. The interference caused by the multi-scale feature to the automatic identification algorithm of the flight stage track deviation distance can be suppressed, and the automatic identification accuracy of the automatic identification of the flight stage track deviation distance can be improved. QAR data can provide accurate real-time track, speed, altitude and other multi-dimensional information of the aircraft during flight, and provide reliable data support for the automatic identification of the direct flight path. By analyzing the track deviation in the QAR data, the method can monitor in real time whether the aircraft is flying according to the predetermined route, thereby realizing accurate evaluation and correction of the track. This method has a high degree of automation, can reduce the risk of human intervention and pilot operation errors, and improve flight safety. Compared with traditional manual judgment or data analysis based on manual records, the direct flight automatic identification method based on QAR data can detect the course deviation in the flight process in real time and accurately, adjust the track in time, and avoid the potential danger of deviation from the route. In addition, this method can efficiently process a large amount of flight data, improve the intelligence level of flight management, and reduce operational complexity. Through data-driven methods, it can not only optimize flight efficiency and reduce fuel consumption, but also improve the accuracy of aircraft heading during flight to ensure flight safety and stability. Therefore, the direct flight automatic identification method based on QAR data has shown obvious advantages in improving flight safety, optimizing flight paths, and improving operational efficiency. BRIEF DESCRIPTION OF THE DRAWINGS

[0040] Figure 1The overall flow chart of the method of the present invention is shown. DETAILED DESCRIPTION

[0041] In order to more clearly understand the above-mentioned purpose, features and advantages of the present invention, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be noted that the embodiments of the present application and the features in the embodiments can be combined with each other without conflict.

[0042] In the following description, many specific details are set forth to facilitate a full understanding of the present invention. However, the present invention may also be implemented in other ways different from those described herein. Therefore, the protection scope of the present invention is not limited to the specific embodiments disclosed below.

[0043] like Figure 1 As shown, the first aspect of the present invention provides a direct flight automatic identification method based on QAR data, comprising the following steps:

[0044] Step S101, establish a QAR data integration center, and use the QAR data integration center to set flight trajectory parameters, obtain the QAR data integration center after the flight trajectory parameters are set, and use the QAR data integration center to control the QAR integrated flight phase position coordinate human-computer interaction information, the flight phase includes take-off phase, climb phase, cruise phase, and descent phase;

[0045] Step S102, using the direct flight information database to integrate the human-computer interaction information of the track deviation distance at different latitudes and longitudes and altitudes during the flight phase, and establishing the t-SNE model perplexity through the human-computer interaction information of the track deviation distance at different latitudes and longitudes and altitudes during the flight phase, and establishing an automatic recognition algorithm for the track deviation distance during the flight phase through the t-SNE model perplexity;

[0046] Step S103, automatically identifying the human-computer interaction information of the position coordinates of the flight stage by using the automatic identification algorithm of the flight stage track deviation distance, obtaining the safety limited interval data within the calibration time of the track deviation distance of the flight stage in the automatic driving state, and judging the attitude stability by using the safety limited interval data within the calibration time of the track deviation distance of the flight stage in the automatic driving state, and generating attitude stability judgment information;

[0047] Step S104: setting a correction scheme for the heading, flight speed, and flight attitude through attitude stability judgment information, and performing aircraft safety control based on the correction scheme for the heading, flight speed, and flight attitude.

[0048] It should be noted that the present invention uses the DE-9IM model to process the target track deviation distance calibration time safety limit interval in the t-SNE model perplexity, so that the input dimension is concentrated in the target track deviation distance calibration time safety limit interval in the t-SNE model perplexity, which can suppress the interference of multi-scale features on the automatic identification algorithm of the flight stage track deviation distance, and can improve the automatic identification accuracy of the automatic identification of the flight stage track deviation distance.

[0049] Furthermore, in the present method, a QAR data integration center is established, and flight trajectory parameters are set for the QAR data integration center to obtain the QAR data integration center after the flight trajectory parameters are set, specifically including:

[0050] Establish a QAR data integration center, and test the QAR data integration center to obtain the ADS-B direct flight parameter jump interval characteristics at different latitudes, longitudes and altitudes of the QAR data integration center, and establish an ADS-B direct flight parameter boundary threshold evaluation model based on the track monitoring system;

[0051] Input the ADS-B direct flight parameter jump interval characteristics of different latitudes, longitudes and altitudes of the QAR data integration center into the ADS-B direct flight parameter boundary threshold evaluation model for training, and obtain the trained ADS-B direct flight parameter boundary threshold evaluation model;

[0052] Obtain the ADS-B direct flight parameter jump interval characteristics of the QAR data integration center within the preset time and input them into the trained ADS-B direct flight parameter boundary threshold evaluation model for evaluation, and obtain the ADS-B direct flight parameter boundary threshold of the QAR data integration center at a unit timestamp;

[0053] The real-time ADS-B direct flight parameters of the QAR data integration center are obtained. When the real-time ADS-B direct flight parameters are greater than the ADS-B direct flight parameter boundary threshold of the QAR data integration center at the unit timestamp, the real-time ADS-B direct flight parameters are adjusted according to the ADS-B direct flight parameter boundary threshold of the QAR data integration center at the unit timestamp to obtain the QAR data integration center after the flight trajectory parameters are set.

[0054] It should be noted that, in fact, since there is a safety limit interval for information transmission, the ADS-B direct flight parameters of the QAR data integration center have a safety limit interval; since the QAR data integration center uses multiple devices for control, the performance of the terminal equipment will degrade to a certain extent after a certain number of years of use (such as a decrease in the amount of information transmitted per unit time), which leads to a safety limit interval for the ADS-B direct flight parameters of the QAR data integration center. The method can adjust the real-time ADS-B direct flight parameters through the ADS-B direct flight parameter boundary threshold of the QAR data integration center at the unit timestamp, so that the human-machine QAR data integration center controls the ADS-B direct flight parameters to meet the predetermined requirements and ensure the stability of the QAR data integration center. QAR (Quick Access Recorder) is a data recording device on an aircraft, mainly used to collect, store and provide various flight data generated during the flight. The design purpose of QAR is to facilitate airlines and related agencies to quickly access the flight data of the aircraft for performance monitoring, fault diagnosis, maintenance and accident investigation.

[0055] Furthermore, in this method, the direct flight information database is used to integrate the human-computer interaction information of the track deviation distance at different latitudes and longitudes and altitudes in the flight phase, and the t-SNE model perplexity is established through the human-computer interaction information of the track deviation distance at different latitudes and longitudes and altitudes in the flight phase. The automatic recognition algorithm of the track deviation distance in the flight phase is established through the t-SNE model perplexity, which specifically includes:

[0056] The direct flight information database is used to integrate the human-computer interaction information of the track deviation distance at different latitudes, longitudes and altitudes during the flight phase, and the track deviation distance type is classified using the human-computer interaction information of the track deviation distance at different latitudes, longitudes and altitudes during the flight phase, and the human-computer interaction information of each track deviation distance type is obtained. The t-SNE model perplexity of each track deviation distance type is established through the human-computer interaction information of each track deviation distance type;

[0057] The human-computer interaction information in the t-SNE model perplexity of each track deviation distance type is input into the safety limit interval clustering calculation model, the safety limit interval within the track deviation distance calibration time existing in the human-computer interaction information per unit time is obtained, and it is determined whether the safety limit interval within the track deviation distance calibration time existing in the human-computer interaction information is equal to or exceeds the preset interval, and the DE-9IM model is introduced;

[0058] When the safety limit interval within the track deviation distance calibration time existing in the human-computer interaction information is equal to or exceeds the preset interval, the safety limit interval within the track deviation distance calibration time corresponding to the safety limit interval information of the track deviation distance type of the unit time human-computer interaction information is obtained as the relevant safety limit interval within the track deviation distance calibration time; when the safety limit interval within the track deviation distance calibration time existing in the human-computer interaction information is less than the preset interval, the safety limit interval within the track deviation distance calibration time corresponding to the safety limit interval information of the track deviation distance type of the unit time human-computer interaction information is used as the relevant safety limit interval within the track deviation distance calibration time;

[0059] The relevant track deviation distance calibration time safety limit interval is input into the DE-9IM model, so that the input dimension is concentrated in the relevant track deviation distance calibration time safety limit interval, the preset position node is obtained, and the flight phase track deviation distance automatic identification algorithm is established based on the route curvature. The preset position node in the t-SNE model perplexity is input into the flight phase track deviation distance automatic identification algorithm for deviation warning.

[0060] The waypoints given by the computer flight plan trajectory are projected to the map coordinate system, and circular buffers are drawn for all waypoints according to the maximum offset distance of the route given by the Civil Aviation Administration. The actual flight trajectory of the QAR is projected to the map coordinate system, and the lines connecting the longitude and latitude trajectory points are formed. The DE-9IM performs logical operations to determine whether the spatial relationship between the actual flight trajectory and the circular buffer of the waypoints intersects, and iterates through the circular buffers of all waypoints to determine whether some waypoints are not passed, and the number of consecutive points not passed is greater than or equal to 1, it is considered that the flight trajectory deviates;

[0061] Extract the starting waypoint and ending waypoint of the flight plan from which the actual flight trajectory deviates, and find the nearest QAR actual flight trajectory point through Euclidean distance to intercept the actual flight trajectory;

[0062] By using the route curvature calculation method, the flight distance of the actual flight trajectory is compared with the straight-line distance from the starting waypoint to the ending waypoint. If the curvature percentage is less than a given threshold, it is determined whether it is a direct flight. The threshold can be set by yourself, and the saved distance is obtained by subtracting the flight distance of the actual flight trajectory from the flight distance of the computer flight plan; Route curvature (%) = (flight distance of actual flight trajectory /

[0063] The straight-line distance from the starting waypoint to the ending waypoint) × 100%.

[0064] It should be noted that in this method, the position coordinate human-computer interaction information of the flight stage is automatically identified by using the automatic identification algorithm of the flight stage track deviation distance, and the safe limited interval data of the flight stage track deviation distance calibration time in the automatic driving state is obtained. The attitude stability is judged by the safe limited interval data of the flight stage track deviation distance calibration time in the automatic driving state, and the attitude stability judgment information is generated, which specifically includes:

[0065] The human-computer interaction information of the position coordinates of the flight phase is input into the automatic identification algorithm of the track deviation distance of the flight phase for automatic identification, and the data of the safe limited interval within the calibration time of the track deviation distance of the flight phase in the automatic driving state is obtained, and the attitude stability judgment deflection angle standard is preset;

[0066] The attitude stability judgment declination standard is used to divide the attitude stability judgment declination data of the safe limited interval within the track deviation distance calibration time during the flight phase under the automatic driving state, obtain the attitude stability judgment declination of the safe limited interval within the track deviation distance calibration time during the flight phase under the automatic driving state, and integrate the navigation deviation information with different magnetic declination angles during the flight phase;

[0067] The attitude stability judgment is generated by judging the declination of the attitude stability within the safe limited interval within the calibration time of the track deviation distance during the flight phase in the automatic driving state, the navigation deviation information with different magnetic declination angles during the flight phase, and the data of the safe limited interval within the calibration time of the track deviation distance during the flight phase in the automatic driving state, and the attitude stability judgment information is generated.

[0068] It should be noted that the standard for attitude stability judgment deflection angle can be set by the type of track deviation distance and the size of the track deviation distance, wherein the attitude stability judgment deflection angle includes low attitude stability judgment deflection angle, medium attitude stability judgment deflection angle, high attitude stability judgment deflection angle, etc.

[0069] Furthermore, in this method, the correction scheme of heading, flight speed and flight attitude is set by judging the attitude stability, which specifically includes:

[0070] The navigation deviation information of different magnetic declination angles in the flight phase is integrated through attitude stability judgment information, and the fixed magnetic declination influence position is established based on the navigation deviation information of different magnetic declination angles in the flight phase. The fixed point correction is performed through the fixed magnetic declination influence position to integrate the track deviation of the flight phase;

[0071] The flight control system error is obtained through the track deviation during the flight phase, and the flight control system error risk coefficient is established. The flight control system error is input into the flight control system error risk coefficient for visualization to obtain risk assessment results of different levels;

[0072] Based on the risk assessment results of different levels, correction plans for heading, flight speed and flight attitude during direct flight under different magnetic declination angles are set, and the correction plans for heading, flight speed and flight attitude are output.

[0073] It should be noted that, since the frequency of turbulence caused by different types of airflow may be limited, this method can be used to formulate a more reasonable correction plan for heading, flight speed, and flight attitude.

[0074] Furthermore, in this method, the aircraft safety control is performed based on the correction scheme of heading, flight speed and flight attitude, specifically including:

[0075] Obtain the distribution characteristic information of turbulence frequency of different airflow types per unit time and the navigation deviation information of different magnetic declination angles in a specific area, and calculate the Cronbach's alpha coefficient between the areas where the turbulence frequency of different airflow types is located and the navigation deviation information of different magnetic declination angles in the flight stage through the distribution characteristic information of turbulence frequency of different airflow types per unit time and the navigation deviation information of different magnetic declination angles in the flight stage;

[0076] Generative adversarial networks were introduced, and weight matrices were set through the generative adversarial networks. The turbulence frequency data of different airflow types were initialized through the Cronbach's α coefficient between the turbulence frequency areas of different airflow types and the navigation deviation information with different magnetic declinations in the flight phase, and the dangerous calibration areas of the turbulence frequency of each different airflow type were obtained.

[0077] Determine whether the Cronbach's alpha coefficient between the areas where the turbulence frequencies of different airflow types are located and the navigation deviation information with different magnetic declinations in the flight phase is greater than a preset Cronbach's alpha coefficient; when the Cronbach's alpha coefficient between the areas where the turbulence frequencies of different airflow types are located and the navigation deviation information with different magnetic declinations in the flight phase is not greater than the preset Cronbach's alpha coefficient, output the danger calibration area of ​​each different airflow type turbulence frequency;

[0078] When the Cronbach's alpha coefficient between the areas where the turbulence frequencies of different airflow types are located and the navigation deviation information with different magnetic declinations in the flight stages is greater than the preset Cronbach's alpha coefficient, the dangerous calibration areas with different turbulence frequencies of different airflow types will be screened in a focused manner until the Cronbach's alpha coefficient between the areas where the turbulence frequencies of different airflow types are located and the navigation deviation information with different magnetic declinations in the flight stages is no greater than the preset Cronbach's alpha coefficient.

[0079] It should be noted that this method can be used to configure the maintenance flight phase of direct flight under different magnetic declination angles, so that each protection area can respond quickly, and the safety control of aircraft with different airflow types and turbulence frequencies is more reasonable.

[0080] In addition, using the QAR data integration center to control the QAR integrated flight phase position coordinate human-machine interaction information may also include the following steps:

[0081] The QAR data integration center is used to obtain the flight route of the QAR per unit time, and the direct flight information database is used to obtain the position coordinate human-computer interaction feature data information under each magnetic declination influence position, and the position coordinate human-computer interaction feature data information under each magnetic declination influence position is input into the t-SNE model; the magnetic declination influence position is used as the first graph node of the t-SNE model, and the position coordinate human-computer interaction feature data information is used as the second graph node of the t-SNE model, an adjacency matrix is ​​established through the first graph node and the second graph node, and the adjacency matrix is ​​input into the knowledge graph for storage; the QAR within a preset time is obtained. The magnetic deviation of AR's flight path is affected by the preset range, and the magnetic deviation of QAR's flight path is affected by the preset range within the preset time. Position information is input into the knowledge graph to obtain the position coordinate human-computer interaction evaluation feature data information of each timestamp; if the position coordinate human-computer interaction evaluation feature data information is greater than the preset position coordinate human-computer interaction feature threshold, the corresponding time period is used as the danger calibration time period of QAR, and the corresponding time period is used as the danger calibration time period of QAR, and data is collected during the danger calibration time period of QAR to integrate the position coordinate human-computer interaction information of the flight stage.

[0082] In addition, the method may also include the following steps: using the direct flight information database to obtain the position coordinate human-computer interaction feature information of each sensor parameter set under each magnetic deflection influence position and the magnetic deflection influence position of the area where the QAR is located per unit time, and obtaining the position coordinate human-computer interaction feature information of the QAR of each sensor parameter through the position coordinate human-computer interaction feature information of each parameter set under each magnetic deflection influence position and the magnetic deflection influence position of the area where the QAR is located per unit time; obtaining the adjustable range of the sensor parameters of the sensor device of the QAR per unit time, and judging whether there is at least one sensor parameter among the sensor parameters that makes the position coordinate human-computer interaction feature information of the QAR greater than the preset position coordinate human-computer interaction feature threshold; if there is at least one sensor parameter among the sensor parameters When there is a sensor parameter that makes the human-computer interaction feature information of the position coordinate of QAR greater than the data of the preset position coordinate human-computer interaction feature threshold, a sensor parameter that makes the human-computer interaction feature information of the position coordinate of QAR greater than the data of the preset position coordinate human-computer interaction feature threshold will be randomly output and used as the sensor parameter of QAR; if there is no sensor parameter among the sensor parameters that makes the human-computer interaction feature information of the position coordinate of QAR greater than the data of the preset position coordinate human-computer interaction feature threshold, the data collection time period of QAR, the flight route of QAR and the data collection point of QAR are adjusted until there is at least one sensor parameter among the sensor parameters that makes the human-computer interaction feature information of the position coordinate of QAR greater than the data of the preset position coordinate human-computer interaction feature threshold.

[0083] It should be noted that due to the influence of the environment, no matter how the sensing parameters are adjusted, there is no parameter that makes the QAR position coordinate human-computer interaction feature information greater than the preset position coordinate human-computer interaction feature threshold data, so that the predetermined standard human-computer interaction cannot be obtained. The use of this method can further improve the rationality of data collection.

[0084] A second aspect of the present invention provides a direct flight automatic identification system based on QAR data, the system comprising a QAR data calculation module and a flight trajectory information comprehensive supervision module, the QAR data calculation module comprising a flight phase track deviation distance automatic identification and attitude stability judgment method program based on position coordinate monitoring, and when the flight phase track deviation distance automatic identification and attitude stability judgment method program based on position coordinate monitoring is executed by the flight trajectory information comprehensive supervision module, the following steps are implemented:

[0085] Establish a QAR data integration center, and use the QAR data integration center to set flight trajectory parameters, obtain the QAR data integration center after the flight trajectory parameters are set, and use the QAR data integration center to control the position coordinate human-computer interaction information of the QAR integration flight phase;

[0086] The direct flight information database is used to integrate the human-computer interaction information of the track deviation distance at different latitudes, longitudes and altitudes during the flight phase, and the t-SNE model perplexity is established through the human-computer interaction information of the track deviation distance at different latitudes, longitudes and altitudes during the flight phase. The automatic identification algorithm of the track deviation distance during the flight phase is established through the t-SNE model perplexity;

[0087] Automatically identify the human-computer interaction information of the position coordinates during the flight phase using the automatic identification algorithm for the track deviation distance during the flight phase, obtain the safety limit interval data within the calibration time of the track deviation distance during the flight phase under the automatic driving state, and judge the attitude stability through the safety limit interval data within the calibration time of the track deviation distance during the flight phase under the automatic driving state, and generate attitude stability judgment information;

[0088] The correction plan for heading, flight speed and flight attitude is set through attitude stability judgment information, and the aircraft is safely controlled based on the correction plan for heading, flight speed and flight attitude.

[0089] Furthermore, in the present system, a QAR data integration center is established, and flight trajectory parameters are set for the QAR data integration center to obtain the QAR data integration center after the flight trajectory parameters are set, specifically including:

[0090] Establish a QAR data integration center, and test the QAR data integration center to obtain the ADS-B direct flight parameter jump interval characteristics at different latitudes, longitudes and altitudes of the QAR data integration center, and establish an ADS-B direct flight parameter boundary threshold evaluation model based on the track monitoring system;

[0091] Input the ADS-B direct flight parameter jump interval characteristics of different latitudes, longitudes and altitudes of the QAR data integration center into the ADS-B direct flight parameter boundary threshold evaluation model for training, and obtain the trained ADS-B direct flight parameter boundary threshold evaluation model;

[0092] Obtain the ADS-B direct flight parameter jump interval characteristics of the QAR data integration center within the preset time and input them into the trained ADS-B direct flight parameter boundary threshold evaluation model for evaluation, and obtain the ADS-B direct flight parameter boundary threshold of the QAR data integration center at a unit timestamp;

[0093] The real-time ADS-B direct flight parameters of the QAR data integration center are obtained. When the real-time ADS-B direct flight parameters are greater than the ADS-B direct flight parameter boundary threshold of the QAR data integration center at the unit timestamp, the real-time ADS-B direct flight parameters are adjusted according to the ADS-B direct flight parameter boundary threshold of the QAR data integration center at the unit timestamp to obtain the QAR data integration center after the flight trajectory parameters are set.

[0094] Furthermore, in this system, the direct flight information database is used to integrate the human-computer interaction information of different latitudes, longitudes, and altitudes of the flight stage, and the t-SNE model perplexity is established through the human-computer interaction information of different latitudes, longitudes, and altitudes of the flight stage. The automatic recognition algorithm of the flight stage track deviation distance is established through the t-SNE model perplexity, which specifically includes:

[0095] The direct flight information database is used to integrate the human-computer interaction information of the track deviation distance at different latitudes, longitudes and altitudes during the flight phase, and the track deviation distance type is classified using the human-computer interaction information of the track deviation distance at different latitudes, longitudes and altitudes during the flight phase, and the human-computer interaction information of each track deviation distance type is obtained. The t-SNE model perplexity of each track deviation distance type is established through the human-computer interaction information of each track deviation distance type;

[0096] The human-computer interaction information in the t-SNE model perplexity of each track deviation distance type is input into the safety limit interval clustering calculation model, the safety limit interval within the track deviation distance calibration time existing in the human-computer interaction information per unit time is obtained, and it is determined whether the safety limit interval within the track deviation distance calibration time existing in the human-computer interaction information is equal to or exceeds the preset interval, and the DE-9IM model is introduced;

[0097] When the safety limit interval within the track deviation distance calibration time existing in the human-computer interaction information is equal to or exceeds the preset interval, the safety limit interval within the track deviation distance calibration time corresponding to the safety limit interval information of the track deviation distance type of the unit time human-computer interaction information is obtained as the relevant safety limit interval within the track deviation distance calibration time; when the safety limit interval within the track deviation distance calibration time existing in the human-computer interaction information is less than the preset interval, the safety limit interval within the track deviation distance calibration time corresponding to the safety limit interval information of the track deviation distance type of the unit time human-computer interaction information is used as the relevant safety limit interval within the track deviation distance calibration time;

[0098] The relevant track deviation distance calibration time safety limit interval is input into the DE-9IM model, so that the input dimension is concentrated in the relevant track deviation distance calibration time safety limit interval, the preset position node is obtained, and the flight phase track deviation distance automatic identification algorithm is established based on the route curvature. The preset position node in the t-SNE model perplexity is input into the flight phase track deviation distance automatic identification algorithm for deviation warning.

[0099] Furthermore, in this system, the position coordinate human-computer interaction information of the flight phase is automatically identified by using the flight phase track deviation distance automatic identification algorithm, and the track deviation distance safety limit interval data of the flight phase in the automatic driving state within the calibration time is obtained. The attitude stability is judged by the track deviation distance safety limit interval data of the flight phase in the automatic driving state within the calibration time, and the attitude stability judgment information is generated, which specifically includes:

[0100] The human-computer interaction information of the position coordinates of the flight phase is input into the automatic identification algorithm of the track deviation distance of the flight phase for automatic identification, and the data of the safe limited interval within the calibration time of the track deviation distance of the flight phase in the automatic driving state is obtained, and the attitude stability judgment deflection angle standard is preset;

[0101] The attitude stability judgment declination standard is used to divide the attitude stability judgment declination data of the safe limited interval within the track deviation distance calibration time during the flight phase under the automatic driving state, obtain the attitude stability judgment declination of the safe limited interval within the track deviation distance calibration time during the flight phase under the automatic driving state, and integrate the navigation deviation information with different magnetic declination angles during the flight phase;

[0102] The attitude stability judgment is generated by judging the declination of the attitude stability within the safe limited interval within the calibration time of the track deviation distance during the flight phase in the automatic driving state, the navigation deviation information with different magnetic declination angles during the flight phase, and the data of the safe limited interval within the calibration time of the track deviation distance during the flight phase in the automatic driving state, and the attitude stability judgment information is generated.

[0103] Furthermore, in this system, the correction scheme of heading, flight speed and flight attitude is set by judging the attitude stability information, which specifically includes:

[0104] The navigation deviation information of different magnetic declination angles in the flight phase is integrated through attitude stability judgment information, and the fixed magnetic declination influence position is established based on the navigation deviation information of different magnetic declination angles in the flight phase. The fixed point correction is performed through the fixed magnetic declination influence position to integrate the track deviation of the flight phase;

[0105] The flight control system error is obtained through the track deviation during the flight phase, and the flight control system error risk coefficient is established. The flight control system error is input into the flight control system error risk coefficient for visualization to obtain risk assessment results of different levels;

[0106] Based on the risk assessment results of different levels, correction plans for heading, flight speed and flight attitude during direct flight under different magnetic declination angles are set, and the correction plans for heading, flight speed and flight attitude are output.

[0107] Furthermore, in this system, the aircraft safety control is performed based on the correction scheme of heading, flight speed and flight attitude, specifically including:

[0108] Obtain the distribution characteristic information of turbulence frequency of different airflow types per unit time and the navigation deviation information of different magnetic declination angles in a specific area, and calculate the Cronbach's alpha coefficient between the areas where the turbulence frequency of different airflow types is located and the navigation deviation information of different magnetic declination angles in the flight stage through the distribution characteristic information of turbulence frequency of different airflow types per unit time and the navigation deviation information of different magnetic declination angles in the flight stage;

[0109] Generative adversarial networks were introduced, and weight matrices were set through the generative adversarial networks. The turbulence frequency data of different airflow types were initialized through the Cronbach's α coefficient between the turbulence frequency areas of different airflow types and the navigation deviation information with different magnetic declinations in the flight phase, and the dangerous calibration areas of the turbulence frequency of each different airflow type were obtained.

[0110] Determine whether the Cronbach's alpha coefficient between the areas where the turbulence frequencies of different airflow types are located and the navigation deviation information with different magnetic declinations in the flight phase is greater than a preset Cronbach's alpha coefficient; when the Cronbach's alpha coefficient between the areas where the turbulence frequencies of different airflow types are located and the navigation deviation information with different magnetic declinations in the flight phase is not greater than the preset Cronbach's alpha coefficient, output the danger calibration area of ​​each different airflow type turbulence frequency;

[0111] When the Cronbach's alpha coefficient between the areas where the turbulence frequencies of different airflow types are located and the navigation deviation information with different magnetic declinations in the flight stages is greater than the preset Cronbach's alpha coefficient, the dangerous calibration areas with different turbulence frequencies of different airflow types will be screened in a focused manner until the Cronbach's alpha coefficient between the areas where the turbulence frequencies of different airflow types are located and the navigation deviation information with different magnetic declinations in the flight stages is no greater than the preset Cronbach's alpha coefficient.

[0112] The third aspect of the present invention provides a computer-readable storage medium, and the computer-readable QAR data calculation module medium includes a method program for automatically identifying the track deviation distance and judging the attitude stability during the flight phase based on position coordinate monitoring. When the method program for automatically identifying the track deviation distance and judging the attitude stability during the flight phase based on position coordinate monitoring is executed by the flight trajectory information comprehensive supervision module, any step of the method for automatically identifying the track deviation distance and judging the attitude stability during the flight phase based on position coordinate monitoring is implemented.

[0113] The above are only specific embodiments of the present invention, but the protection scope of the present invention is not limited thereto. Any person skilled in the art who is familiar with the technical field can easily think of changes or substitutions within the technical scope disclosed by the present invention, which should be included in the protection scope of the present invention. Therefore, the protection scope of the present invention should be based on the protection scope of the claims.

Claims

1. A direct flight automatic identification method based on QAR data, characterized in that: The following steps are involved: Step S101, establishing a QAR data integration center, and using the QAR data integration center to set flight trajectory parameters, obtaining the QAR data integration center after the flight trajectory parameters are set, and using the QAR data integration center to control the QAR integrated flight phase position coordinate human-computer interaction information, the flight phase includes take-off phase, climb phase, cruise phase, and descent phase; Step S102, using the direct flight information database to integrate the human-computer interaction information of the different longitudes and latitudes and altitudes of the flight stage, and establishing the t-SNE model perplexity through the human-computer interaction information of the different longitudes and latitudes and altitudes of the flight stage, and establishing the flight stage track deviation distance automatic recognition algorithm through the t-SNE model perplexity; Step S103, automatically identifying the position coordinate human-computer interaction information of the flight stage by using the automatic identification algorithm for the flight stage track deviation distance, obtaining the safety limited interval data within the calibration time of the track deviation distance of the flight stage in the automatic driving state, and performing attitude stability judgment based on the safety limited interval data within the calibration time of the track deviation distance of the flight stage in the automatic driving state, and generating attitude stability judgment information; Step S104: setting a correction scheme for heading, flight speed, and flight attitude based on the attitude stability judgment information, and performing aircraft safety control based on the correction scheme for heading, flight speed, and flight attitude.

2. The method for automatic direct flight identification based on QAR data according to claim 1, characterized in that: The direct flight information database is used to integrate the human-computer interaction information of different longitudes and latitudes, altitudes and track deviation distances in the flight phase, and the t-SNE model perplexity is established through the human-computer interaction information of different longitudes and latitudes, altitudes and track deviation distances in the flight phase. The automatic recognition algorithm of the track deviation distance in the flight phase is established through the t-SNE model perplexity, which specifically includes: The direct flight information database is used to integrate the human-computer interaction information of the track deviation distance at different latitudes, longitudes and altitudes in the flight phase, and the track deviation distance type is classified by using the human-computer interaction information of the track deviation distance at different latitudes, longitudes and altitudes in the flight phase, and the human-computer interaction information of each track deviation distance type is obtained, and the t-SNE model perplexity of each track deviation distance type is established through the human-computer interaction information of each track deviation distance type; The human-computer interaction information in the t-SNE model perplexity of each track deviation distance type is input into the safety limit interval clustering calculation model, the safety limit interval within the track deviation distance calibration time existing in the human-computer interaction information per unit time is obtained, and it is determined whether the safety limit interval within the track deviation distance calibration time existing in the human-computer interaction information is equal to or exceeds the preset interval, and the DE-9IM model is introduced; If the safety limit interval within the track deviation distance calibration time existing in the human-computer interaction information is equal to or exceeds the preset interval, the safety limit interval within the track deviation distance calibration time corresponding to the safety limit interval information of the track deviation distance type of the unit time human-computer interaction information is obtained as the relevant safety limit interval within the track deviation distance calibration time; if the safety limit interval within the track deviation distance calibration time existing in the human-computer interaction information is less than the preset interval, the safety limit interval within the track deviation distance calibration time corresponding to the safety limit interval information of the track deviation distance type of the unit time human-computer interaction information is used as the relevant safety limit interval within the track deviation distance calibration time; Input the relevant track deviation distance calibration time safety limit interval into the DE-9IM model, so that the input dimension is concentrated in the relevant track deviation distance calibration time safety limit interval, obtain the preset position node, establish the flight stage track deviation distance automatic identification algorithm based on the route curvature, and input the preset position node into the flight stage track deviation distance automatic identification algorithm for deviation warning; The waypoints given by the computer flight plan trajectory are projected to the map coordinate system, and a circular buffer is drawn for all waypoints according to the maximum offset distance of the route given by the Civil Aviation Administration. The actual flight trajectory of the QAR is projected to the map coordinate system, and a line connecting the longitude and latitude trajectory points is formed. The DE-9IM performs a logical operation to determine whether the spatial relationship between the actual flight trajectory and the circular buffer of the waypoints intersects, and iterates through the circular buffer of the right waypoints. If some waypoints are not passed, and the number of consecutive points not passed is greater than or equal to 1, it is considered that the flight trajectory deviates; Extract the starting waypoint and ending waypoint of the flight plan from which the actual flight trajectory deviates, and find the nearest QAR actual flight trajectory point through Euclidean distance to intercept the actual flight trajectory; By adopting the calculation method of route curvature, comparing the flight distance of the actual flight trajectory with the straight-line distance from the starting waypoint to the ending waypoint, if the curvature percentage is less than a given threshold, it is determined whether it is a direct flight. The threshold can be set by itself, and the saved distance is obtained by subtracting the flight distance of the actual flight trajectory from the flight distance of the computer flight plan; route curvature (%) = (flight distance of the actual flight trajectory ÷ straight-line distance from the starting waypoint to the ending waypoint) × 100%.

3. The method for automatic direct flight identification based on QAR data according to claim 1, characterized in that: Establishing a QAR data integration center, and using the QAR data integration center to set flight trajectory parameters, and obtaining the QAR data integration center after the flight trajectory parameters are set, specifically includes: Establish a QAR data integration center, and test the QAR data integration center to obtain the ADS-B direct flight parameter jump interval characteristics at different latitudes, longitudes and altitudes of the QAR data integration center, and establish an ADS-B direct flight parameter boundary threshold evaluation model based on the track monitoring system; Inputting the ADS-B direct flight parameter jump interval characteristics of different latitudes, longitudes and altitudes of the QAR data integration center into the ADS-B direct flight parameter boundary threshold evaluation model for training, and obtaining the trained ADS-B direct flight parameter boundary threshold evaluation model; Obtaining the ADS-B direct flight parameter jump interval characteristics of the QAR data integration center within a preset time and inputting them into the trained ADS-B direct flight parameter boundary threshold evaluation model for evaluation, and obtaining the ADS-B direct flight parameter boundary threshold of the QAR data integration center at a unit timestamp; The real-time ADS-B direct flight parameter of the QAR data integration center is obtained. If the real-time ADS-B direct flight parameter is greater than the ADS-B direct flight parameter boundary threshold of the QAR data integration center at the unit timestamp, the real-time ADS-B direct flight parameter is adjusted according to the ADS-B direct flight parameter boundary threshold of the QAR data integration center at the unit timestamp to obtain the QAR data integration center after the flight trajectory parameters are set.

4. The method for automatic direct flight identification based on QAR data according to claim 1, characterized in that: The position coordinate human-computer interaction information of the flight phase is automatically identified by using the automatic identification algorithm for the flight phase track deviation distance, and the safety limited interval data within the calibration time of the track deviation distance of the flight phase in the automatic driving state is obtained. The attitude stability is judged by using the safety limited interval data within the calibration time of the track deviation distance of the flight phase in the automatic driving state, and the attitude stability judgment information is generated, which specifically includes: Inputting the position coordinate human-machine interaction information of the flight phase into the automatic identification algorithm of the track deviation distance of the flight phase for automatic identification, obtaining the safety limit interval data within the calibration time of the track deviation distance of the flight phase in the automatic driving state, and presetting the attitude stability judgment deflection angle standard; The attitude stability judgment declination standard is used to divide the data of the safe limited interval within the track deviation distance calibration time during the flight phase in the automatic driving state into attitude stability judgment declination, obtain the attitude stability judgment declination within the safe limited interval within the track deviation distance calibration time during the flight phase in the automatic driving state, and integrate the navigation deviation information with different magnetic declination angles during the flight phase; The attitude stability judgment is generated by judging the declination of the attitude stability in the safety limited area within the calibration time of the track deviation distance during the flight phase in the automatic driving state, the navigation deviation information with different magnetic declination angles during the flight phase, and the data of the safety limited area within the calibration time of the track deviation distance during the flight phase in the automatic driving state, and the attitude stability judgment information is generated.

5. The method for automatic direct flight identification based on QAR data according to claim 1, characterized in that: The correction scheme of heading, flight speed and flight attitude is set according to the attitude stability judgment information, specifically including: Integrate navigation deviation information of different magnetic declination angles in flight phases through the attitude stability judgment information, establish a fixed-point magnetic declination influence position based on the navigation deviation information of different magnetic declination angles in the flight phase, perform fixed-point correction through the fixed-point magnetic declination influence position, and integrate the track deviation of the flight phase; Acquire a flight control system error through the track deviation of the flight phase, establish a flight control system error risk coefficient, input the flight control system error into the flight control system error risk coefficient for visualization, and obtain risk assessment results of different levels; According to the risk assessment results of different levels, correction schemes for the heading, flight speed and flight attitude of direct flight under different magnetic declination angles are set, and the correction schemes for the heading, flight speed and flight attitude are output.

6. The method for automatic direct flight identification based on QAR data according to claim 1, characterized in that: The aircraft is safely controlled according to the correction scheme of the heading, flight speed and flight attitude, including: Obtaining distribution characteristic information of turbulence frequencies of different airflow types per unit time and navigation deviation information of different magnetic declination angles in a specific area, and calculating the Cronbach's alpha coefficient between the areas where the turbulence frequencies of different airflow types are located and the navigation deviation information of different magnetic declination angles in the flight stage through the distribution characteristic information of turbulence frequencies of different airflow types per unit time and the navigation deviation information of different magnetic declination angles in the flight stage; A generative adversarial network is introduced, a weight matrix is ​​set by the generative adversarial network, and the turbulence frequency data of different airflow types are initialized by the Cronbach's alpha coefficient between the areas where the turbulence frequencies of different airflow types are located and the navigation deviation information with different magnetic declinations in the flight phase, so as to obtain the dangerous calibration area of ​​the turbulence frequency of each different airflow type; Determine whether the Cronbach's alpha coefficient between the areas where the turbulence frequencies of different airflow types are located and the navigation deviation information with different magnetic declinations in the flight phase is greater than a preset Cronbach's alpha coefficient; if the Cronbach's alpha coefficient between the areas where the turbulence frequencies of different airflow types are located and the navigation deviation information with different magnetic declinations in the flight phase is not greater than the preset Cronbach's alpha coefficient, output the danger calibration area of ​​each different airflow type turbulence frequency; If the Cronbach's alpha coefficient between the areas where the turbulence frequencies of different airflow types are located and the navigation deviation information with different magnetic declinations in the flight stages is greater than the preset Cronbach's alpha coefficient, the dangerous calibration areas for each different airflow type turbulence frequency are screened in a focused manner until the Cronbach's alpha coefficient between the areas where the turbulence frequencies of different airflow types are located and the navigation deviation information with different magnetic declinations in the flight stages is not greater than the preset Cronbach's alpha coefficient.

Images