Flight cruise state recognition processing method and system based on multi-source data

Abstract

This invention discloses a method and system for identifying and processing flight cruise status based on multi-source data. The method includes: S1, constructing a planned cruise status identification model and training it for planned cruise status feature recognition by combining it with flight characteristics; S2, constructing an actual cruise status identification model and training it for actual cruise status feature recognition by combining it with flight characteristics; S3, performing feature association mapping between the planned cruise status identification model and the actual cruise status identification model according to cruise phases; S4, collecting current flight data and inputting it into the actual cruise status identification model for feature recognition and extraction, and using the feature association mapping between the planned cruise status identification model and the actual cruise status identification model to calibrate and identify whether the flight is currently in cruise status, cruise phase, and cruise phase start and end point data. This invention achieves real-time detection of aircraft cruise status relying on multi-source data, improving detection efficiency and accuracy.

Description

Technical Field

[0001] This invention relates to the field of civil aviation flight status monitoring, and in particular to a method and system for identifying and processing flight cruise status based on multi-source data. Background Technology

[0002] In the civil aviation sector, when a civil aircraft is in cruise mode, it exhibits advantages in multiple aspects, playing a crucial supporting role in the efficient and safe operation of the aviation industry. From a fuel efficiency perspective, during cruise, the aircraft reaches a specific altitude and maintains a relatively stable speed, at which point its aerodynamic performance is optimized. For example, cruising at the bottom of the stratosphere, where the air is thin, the air resistance encountered by the aircraft is significantly reduced. Compared to low-altitude flight, the engine does not need to output excessive power to overcome strong drag, and fuel consumption is effectively controlled. In terms of flight stability, the cruise altitude environment is relatively stable with less airflow interference. The aircraft can maintain a stable flight attitude, reducing turbulence, greatly improving passenger comfort, and reducing the workload and safety risks for cabin crew. At the same time, stable flight helps pilots accurately maintain flight parameters, reducing operational burden and allowing them to focus more on monitoring flight status and responding to emergencies, thus enhancing flight safety. In terms of balancing speed and range, flying at a moderate cruise speed allows the aircraft to reach its destination efficiently while avoiding excessive wear and tear on the airframe and fuel consumption. In summary, the fuel efficiency, flight stability, and speed-range balance advantages of civil aircraft during the cruise phase collectively shape the efficient, safe, and comfortable service quality of civil aviation transportation, becoming an indispensable key element for the vigorous development of the aviation industry. Currently, determining whether a civil aircraft has entered the cruise phase mainly relies on human experience, and there are no effective detection methods. To improve situational awareness and risk factor analysis during flight operations, and to implement more refined flight operation monitoring, it is necessary to effectively identify and detect flight nodes that indicate the aircraft has entered the cruise phase, thereby significantly improving the safety margin of flight operations. Summary of the Invention

[0003] The purpose of this invention is to provide a flight cruise state identification and processing method based on multi-source data. It realizes the feature association and mapping between the planned cruise state identification model and the actual cruise state identification model according to the cruise stage, and realizes the mutual judgment and feature association between the aircraft's planned and actual flight in the aircraft cruise state. Based on the data collection of the aircraft at the current time, it can realize the real-time and accurate judgment of whether the aircraft has entered the cruise state.

[0004] The objective of this invention is achieved through the following technical solution:

[0005] A method for identifying and processing flight cruise status based on multi-source data, the method comprising:

[0006] S1. Collect historical aircraft plan data of the aircraft being studied to form an aircraft plan database. Extract the same aircraft plan data for the aircraft and flight being studied from the aircraft plan database. The aircraft plan data includes planned takeoff time, planned landing time, planned waypoint, and planned cruising altitude. Construct a planned cruise state recognition model and combine it with flight characteristics to train the planned cruise state feature recognition using the aircraft plan database.

[0007] S2. Collect historical flight dynamics data and historical ADS-B data of the aircraft to form an actual flight database, which includes altitude and rate of climb / drop; construct an actual cruise state recognition model and combine it with flight characteristics to train the actual cruise state feature recognition using the actual flight database.

[0008] S3. The planned cruise status identification model obtains the planned cruise start-point features and planned cruise end-point features according to the cruise phase, and the actual cruise status identification model obtains the actual cruise start-point features and actual cruise end-point features according to the cruise phase. The planned cruise status identification model and the actual cruise status identification model are mapped by feature association according to the cruise phase.

[0009] S4. Collect the flight dynamic data and ADS-B data of the current flight status and input them into the actual cruise status identification model for feature identification and extraction. Use the feature association mapping between the planned cruise status identification model and the actual cruise status identification model to calibrate and identify whether the flight is in cruise status, cruise phase and cruise phase start and end point data. Cruise phase start and end point data include cruise phase information, start altitude and time, and end altitude and time.

[0010] To better achieve the present invention, the present invention also includes the following methods:

[0011] S5. For flights that have not reached cruise status, output the predicted cruise phase start and end point data. The predicted cruise phase start and end point data includes the predicted cruise phase information, the predicted start altitude and time, and the predicted end altitude and time.

[0012] Preferably, the features identified by the planned cruise state identification model include the total planned flight duration, the flight climb peak (TOC), the planned cruise altitude, the planned cruise waypoints, and the simulated cruise waypoints. The simulated cruise waypoints are cruise start and end waypoints simulated according to the cruise phase by using the feature association mapping between the planned cruise state identification model and the actual cruise state identification model.

[0013] Preferably, the features identified by the actual cruise state identification model include the total planned flight duration, altitude, rate of ascent and descent, and duration. The method for associating and mapping the features of the planned cruise state identification model with those of the actual cruise state identification model is as follows:

[0014] If the actual cruise status identification model identifies the total planned flight time as greater than H1 and the altitude as greater than M1, then the associated features of the actual cruise status identification model and the planned cruise status identification model are identified as cruise status.

[0015] If the actual cruise state identification model identifies an ascent / descent rate less than V1 and a duration of H2, then the associated features of the actual cruise state identification model and the planned cruise state identification model are identified as cruise states.

[0016] If the altitude identified by the actual cruise state identification model is greater than the flight climb peak (TOC) of the planned cruise state identification model, then the associated features of the actual cruise state identification model and the planned cruise state identification model are identified as cruise state.

[0017] Preferably, the feature association mapping method between the planned cruise state identification model and the actual cruise state identification model further includes: if the altitude identified by the actual cruise state identification model minus the flight climb peak TOC value of the planned cruise state identification model is greater than the threshold A1 and the duration is H3, then the associated features of the actual cruise state identification model and the planned cruise state identification model are identified as cruise state.

[0018] Preferably, the H1 setting range is 85-98 minutes, the M1 setting range is 5995-6010 meters, the V1 setting range is 95-110 m / min, and the H2 setting range is 10-20 minutes.

[0019] A method for identifying and processing flight cruise status based on multi-source data includes a data acquisition module, an aircraft plan database, an actual aircraft flight database, a planned cruise status identification model, an actual cruise status identification model, and a feature association mapping processing model. The aircraft plan database stores historical aircraft plan data for the flight under study. The planned cruise status identification model extracts planned flight plan data for the same aircraft and flight from the aircraft plan database. This planned flight plan data includes planned takeoff time, planned landing time, planned waypoint, and planned cruise altitude. The planned cruise status identification model combines flight characteristics with feature extraction from the aircraft plan database and obtains planned cruise start-point and planned cruise end-point features according to the cruise phases. The actual aircraft flight database stores historical flight dynamic data and historical ADS-B data for the flight under study. Both flight dynamic data and historical ADS-B data include altitude and rate of ascent / descent. The actual cruise status identification model combines flight characteristics with the aircraft's actual flight database to extract features and obtain actual cruise start-point and actual cruise end-point features according to the cruise phase. The feature association mapping processing model maps the planned cruise status identification model and the actual cruise status identification model according to the cruise phase and constructs a feature association relationship library. The data acquisition module collects the flight dynamic data and ADS-B data of the aircraft at the current moment and inputs them into the actual cruise status identification model for feature identification and extraction. Using the feature association mapping between the planned cruise status identification model and the actual cruise status model, the model calibrates and identifies whether the aircraft is currently in cruise status, cruise phase, and cruise phase start-end-end data. Cruise phase start-end-end data includes cruise phase information, start-point altitude and time, and end-point altitude and time.

[0020] Compared with the prior art, the present invention has the following advantages and beneficial effects:

[0021] This invention constructs a planned cruise state recognition model and trains it using an aircraft plan database for cruise state-related feature recognition. It also constructs an actual cruise state recognition model and trains it using an actual aircraft flight database for cruise state-related feature recognition. For the same aircraft and flight, it achieves feature association mapping between the planned and actual cruise state recognition models according to the cruise phase. This enables the combined judgment and feature association between the aircraft plan and the actual flight during the aircraft's cruise state. Based on the data collection of the aircraft at the current moment, it can realize real-time and accurate judgment of whether the aircraft has entered the cruise state. It realizes real-time detection of the aircraft's cruise state relying on multi-source data, improving detection efficiency and accuracy. Attached Figure Description

[0022] Figure 1 This is a flowchart of the method of the present invention. Detailed Implementation

[0023] The present invention will be further described in detail below with reference to embodiments: Example

[0024] like Figure 1 As shown, a flight cruise status identification and processing method based on multi-source data is described, the method comprising:

[0025] S1. Collect historical aircraft planning data of the aircraft being studied to form an aircraft planning database. Extract the same aircraft planning data of the aircraft and flight being studied from the aircraft planning database. The aircraft planning data includes planned takeoff time, planned landing time, planned waypoints, planned cruising altitude, and at least the total planned flight duration, flight climb peak (TOC), planned cruising altitude, planned cruising waypoints, and simulated cruising waypoints. A planned cruise state recognition model is constructed and trained using an aircraft plan database to identify planned cruise state features, based on flight characteristics. After training, the planned cruise state recognition model primarily identifies features such as total planned flight duration, Total Climb Point (TOC), planned cruise altitude, planned cruise waypoints, and simulated cruise waypoints (and may also include features corresponding to other data). The simulated cruise waypoints are cruise start and end point points simulated according to the cruise phase by mapping the features of the planned cruise state recognition model and the actual cruise state recognition model. (These cruise start and end point points are cruise path points simulated by cross-validating the planned cruise state recognition model and the actual cruise state recognition model on the same research aircraft and flight, and are used as cruise start and end point points.)

[0026] S2. Collect historical flight dynamics data and historical ADS-B data of the aircraft to form an actual flight database. The data content of the actual flight database includes altitude and rate of climb / descendancy. The formula for calculating the rate of climb / descendancy is as follows: ,in

[0027] This represents the altitude of the flight trajectory at time t (usually the current time). This represents the altitude of the flight trajectory at time t-1 (generally the time before the current time). This refers to the time difference between two flight trajectory points. An actual cruise state recognition model is constructed, combining flight characteristics with an actual aircraft flight database for training to identify actual cruise state features. After training, the actual cruise state recognition model primarily identifies features such as total planned flight duration, altitude, rate of ascent / descent, and duration. The actual aircraft flight database includes at least data on total planned flight duration, altitude, rate of ascent / descent, and duration; it can also contain other data and extract features corresponding to other data.

[0028] S3. The planned cruise status identification model obtains the planned cruise start-point features and planned cruise end-point features according to the cruise phase. The actual cruise status identification model obtains the actual cruise start-point features and actual cruise end-point features according to the cruise phase. The planned cruise status identification model and the actual cruise status identification model are then mapped to each other according to the cruise phase.

[0029] After training the actual cruise state recognition model, the main features identified by the model are the total planned flight time, altitude, rate of ascent and descent, and duration. The mapping method for the feature association between the planned cruise state recognition model and the actual cruise state recognition model is as follows:

[0030] If the actual cruise status identification model identifies a total planned flight duration greater than H1 and an altitude greater than M1, then the associated features of the actual cruise status identification model and the planned cruise status identification model are identified as cruise status. The preferred range for H1 in this invention is 85–98 minutes, and the preferred range for M1 is 5995–6010 meters. If the actual cruise status identification model identifies a rate of ascent / descent less than V1 and a duration of H2, then the associated features of the actual cruise status identification model and the planned cruise status identification model are identified as cruise status. The preferred range for V1 in this invention is 95–110 m / min, and the preferred range for H2 is 10–20 minutes. If the altitude identified by the actual cruise status identification model is greater than the flight climb peak (TOC) of the planned cruise status identification model, then the associated features of the actual cruise status identification model and the planned cruise status identification model are identified as cruise status. In some embodiments, the feature association mapping method between the planned cruise state identification model and the actual cruise state identification model of the present invention further includes: if the altitude identified by the actual cruise state identification model minus the flight climb peak TOC value of the planned cruise state identification model is greater than the threshold A1 (the threshold A1 in the present invention is selected in the range of -10 to 5, and the threshold A1 in this embodiment is zero), and the duration is H3 (the duration H3 in this embodiment is 10 minutes), then the associated features of the actual cruise state identification model and the planned cruise state identification model are identified as cruise state.

[0031] S4. Collect the flight dynamic data and ADS-B data of the current flight status and input them into the actual cruise status identification model for feature identification and extraction. Use the feature association mapping between the planned cruise status identification model and the actual cruise status identification model to calibrate and identify whether the flight is in cruise status, cruise phase and cruise phase start and end point data. Cruise phase start and end point data include cruise phase information, start altitude and time, and end altitude and time.

[0032] S5. For flights that have not reached cruise status, output the predicted cruise phase start and end point data. The predicted cruise phase start and end point data includes the predicted cruise phase information, the predicted start altitude and time, and the predicted end altitude and time.

[0033] A method for identifying and processing flight cruise status based on multi-source data includes a data acquisition module, an aircraft plan database, an actual aircraft flight database, a planned cruise status identification model, an actual cruise status identification model, and a feature association mapping processing model. The aircraft plan database stores historical aircraft plan data for the flight under study. The planned cruise status identification model extracts planned flight plan data for the same aircraft and flight from the aircraft plan database. This planned flight plan data includes planned takeoff time, planned landing time, planned waypoint, and planned cruise altitude. The planned cruise status identification model combines flight characteristics with feature extraction from the aircraft plan database and obtains planned cruise start-point and planned cruise end-point features according to the cruise phases. The actual aircraft flight database stores historical flight dynamic data and historical ADS-B data for the flight under study. Both flight dynamic data and historical ADS-B data include altitude and rate of ascent / descent. The actual cruise status identification model combines flight characteristics with the aircraft's actual flight database to extract features and obtain actual cruise start-point and actual cruise end-point features according to the cruise phase. The feature association mapping processing model maps the planned cruise status identification model and the actual cruise status identification model according to the cruise phase and constructs a feature association relationship library. The data acquisition module collects the flight dynamic data and ADS-B data of the aircraft at the current moment and inputs them into the actual cruise status identification model for feature identification and extraction. Using the feature association mapping between the planned cruise status identification model and the actual cruise status model, the model calibrates and identifies whether the aircraft is currently in cruise status, cruise phase, and cruise phase start-end-end data. Cruise phase start-end-end data includes cruise phase information, start-point altitude and time, and end-point altitude and time.

[0034] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for identifying and processing flight cruise status based on multi-source data, characterized in that: The methods include: S1. Collect historical aircraft plan data of the aircraft being studied to form an aircraft plan database. Extract the same aircraft plan data for the aircraft and flight being studied from the aircraft plan database. The aircraft plan data includes planned takeoff time, planned landing time, planned waypoint, and planned cruising altitude. Construct a planned cruise state recognition model and combine it with flight characteristics to train the planned cruise state feature recognition using the aircraft plan database. S2. Collect historical flight dynamics data and historical ADS-B data of the aircraft to form an actual flight database, which includes altitude and rate of climb / drop; construct an actual cruise state recognition model and combine it with flight characteristics to train the actual cruise state feature recognition using the actual flight database. S3. The planned cruise status identification model obtains the planned cruise start-point features and planned cruise end-point features according to the cruise phase, and the actual cruise status identification model obtains the actual cruise start-point features and actual cruise end-point features according to the cruise phase. The planned cruise status identification model and the actual cruise status identification model are mapped by feature association according to the cruise phase. S4. Collect the flight dynamic data and ADS-B data of the current flight status and input them into the actual cruise status identification model for feature identification and extraction. Use the feature association mapping between the planned cruise status identification model and the actual cruise status identification model to calibrate and identify whether the flight is in cruise status, cruise phase and cruise phase start and end point data. Cruise phase start and end point data include cruise phase information, start altitude and time, and end altitude and time.

2. The flight cruise status identification and processing method based on multi-source data according to claim 1, characterized in that: It also includes the following methods: S5. For flights that have not reached cruise status, output the predicted cruise phase start and end point data. The predicted cruise phase start and end point data includes the predicted cruise phase information, the predicted start altitude and time, and the predicted end altitude and time.

3. The flight cruise status identification and processing method based on multi-source data according to claim 1, characterized in that: The features identified by the planned cruise state identification model include the total planned flight duration, the flight's climb peak (TOC), the planned cruise altitude, the planned cruise waypoints, and the simulated cruise waypoints. The simulated cruise waypoints are cruise start and end waypoints simulated according to the cruise phase by using the feature association mapping between the planned cruise state identification model and the actual cruise state identification model.

4. The flight cruise status identification and processing method based on multi-source data according to claim 3, characterized in that: The features identified by the actual cruise state identification model include the total planned flight time, altitude, rate of ascent and descent, and duration. The mapping method for associating the features of the planned cruise state identification model and the actual cruise state identification model is as follows: If the actual cruise status identification model identifies the total planned flight time as greater than H1 and the altitude as greater than M1, then the associated features of the actual cruise status identification model and the planned cruise status identification model are identified as cruise status. If the actual cruise state identification model identifies an ascent / descent rate less than V1 and a duration of H2, then the associated features of the actual cruise state identification model and the planned cruise state identification model are identified as cruise states. If the altitude identified by the actual cruise state identification model is greater than the flight climb peak (TOC) of the planned cruise state identification model, then the associated features of the actual cruise state identification model and the planned cruise state identification model are identified as cruise state.

5. The flight cruise status identification and processing method based on multi-source data according to claim 4, characterized in that: The method for mapping the features of the planned cruise state identification model and the actual cruise state identification model also includes: if the difference between the altitude identified by the actual cruise state identification model and the flight climb peak TOC value of the planned cruise state identification model is greater than the threshold A1 and the duration is H3, then the associated features of the actual cruise state identification model and the planned cruise state identification model are identified as cruise state.

6. The flight cruise status identification and processing method based on multi-source data according to claim 4, characterized in that: The H1 setting range is 85–98 minutes, the M1 setting range is 5995–6010 meters, the V1 setting range is 95–110 m / min, and the H2 setting range is 10–20 minutes.

7. A flight cruise status identification and processing system based on multi-source data, characterized in that: It includes a data acquisition module, an aircraft plan database, an actual aircraft flight database, a planned cruise status identification model, an actual cruise status identification model, and a feature association mapping processing model. The aircraft plan database stores historical aircraft plan data of the aircraft being studied. The planned cruise status identification model extracts the same aircraft plan data for the aircraft and flight being studied from the aircraft plan database. The aircraft plan data includes planned takeoff time, planned landing time, planned waypoint, and planned cruise altitude. The planned cruise status identification model combines flight characteristics with the aircraft plan database to extract features and obtain the planned cruise start-point features and planned cruise end-point features according to the cruise phase. The aircraft's actual flight database internally stores historical flight dynamics data and historical ADS-B data for the research flight. Both historical flight dynamics and historical ADS-B data include altitude and rate of climb / drop. The actual cruise state identification model, combined with flight characteristics, utilizes the aircraft's actual flight database to extract features and obtain actual cruise start-point and actual cruise end-point features according to the cruise phase. The feature association mapping processing model maps the planned cruise state identification model and the actual cruise state identification model according to the cruise phase and constructs a feature association database. The data acquisition module collects the flight dynamics data and ADS-B data of the aircraft at the current moment and inputs them into the actual cruise state identification model for feature recognition and extraction. Using the feature association mapping between the planned and actual cruise state identification models, the model calibrates and identifies whether the aircraft is currently in cruise state, cruise phase, and cruise phase start-end and end-end data. Cruise phase start-end and end-end data include cruise phase information, start-point altitude and time, and end-point altitude and time.

Images