A method and system for identifying aircraft turbulence based on multi-source data

Abstract

The application provides a method and system for identifying aircraft turbulence based on multi-source data. The method includes using ADS-B track data and aircraft in-flight reports to locate the position and height of the turbulence occurrence according to the flight number and turbulence time, and determining the initial target turbulence position. Based on QAR data, the EDR index is calculated, the initial target turbulence position located in step S1 is expanded in the horizontal and vertical directions by 600 meters, the turbulence area is identified and output. The turbulence area expanded in step S2, the turbulence area located by AMDAR data, and each turbulence index in step S3 are compared and scored to output a scoring result. Based on the scoring result, the turbulence index is weighted and fused to generate a fused turbulence index. According to the fused turbulence index, the turbulence area is identified. The above method significantly improves the efficiency and accuracy of aircraft turbulence area identification.

Description

Technical Field

[0001] This application relates to the field of aviation safety data identification and analysis technology, and in particular to an aviation turbulence identification method and system based on multi-source data. Background Technology

[0002] Turbulence is one of the core risks threatening flight safety. According to statistics from the International Civil Aviation Organization (ICAO), more than 50,000 turbulence incidents are reported globally annually over the past decade, with 15% resulting in personal injury or aircraft damage. Traditional turbulence identification methods, which rely on pilot voice reports (PIREP), suffer from three major drawbacks: fuzzy positioning (average position deviation ±50km), subjective intensity (e.g., lack of quantitative standards for "moderate turbulence"), and incomplete coverage (reporting rate is less than 30% of actual turbulence). These shortcomings make it difficult to meet the needs of modern aviation safety early warning systems.

[0003] Existing technical solutions and their limitations: For example, the meteorological index method: calculates turbulence indices (such as the Ellrod index and Brownian index) based on numerical forecast grid fields. It relies on the accuracy of meteorological models, resulting in a false alarm rate exceeding 40% under complex weather conditions; it cannot distinguish between mechanical and thermal turbulence, with a false alarm rate of up to 35% in jet stream areas. Airborne sensor method: utilizes AMDAR (Aircraft Weather Data Relay) or QAR (Quick Access Recorder) data to invert turbulence. AMDAR sampling is sparse (global coverage <15% of flight routes) and is only carried out on large passenger aircraft; therefore, how to achieve accurate identification of turbulence areas is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0004] This application provides an aviation turbulence identification method and system based on multi-source data to solve the problems of low efficiency and poor accuracy in turbulence area identification in the prior art.

[0005] Firstly, this application provides an aviation turbulence identification method based on multi-source data, including:

[0006] S1. Using ADS-B track data and aircraft in-flight reports, locate the location and altitude of the turbulence based on the flight number and turbulence time, and determine the initial target turbulence location;

[0007] S2. Calculate the EDR index based on QAR data, and perform a horizontal adjustment on the initial target bump position located in step S1. A vertical area of ​​600 meters is expanded to identify and output bumpy areas;

[0008] S3. Calculate at least several turbulence indices using gridded field data, including Brownian index, MOS CAT probability prediction factor, Dutton index, Ellrod_TI1, Ellrod_TI2, horizontal temperature gradient index, and CCAT index.

[0009] S4. Compare and score the bumpy area expanded in step S2, the bumpy area located by AMDAR data, and each of the bumpy indices in step S3, and output the scoring results. Based on the scoring results, the bumpy indices are weighted and fused to generate a fused bumpy index.

[0010] S5. Identify the bumpy area based on the fused bump index.

[0011] Preferably, as one possible implementation: S11, obtain the flight number, and parse the timestamp, longitude, latitude, and altitude fields of the corresponding ADS-B waypoint according to the flight number; S12, match the turbulence occurrence time in the aircraft's in-flight report according to the ADS-B waypoint, and determine the initial target turbulence location.

[0012] Preferably, as one possible implementation; step S12, matching the turbulence occurrence time in the aircraft's in-flight report and determining the initial target turbulence location, includes:

[0013] S121. Align the timestamps in the aircraft's in-flight reports with the data timeline of the ADS-B trackpoints;

[0014] S122. Select ADS-B waypoints within ±30 seconds before and after the report time;

[0015] S123. If there are multiple ADS-B track points, select the geometric center coordinates as the positioning reference point; if there is only one ADS-B track point, determine that track point as the positioning reference point, determine whether the in-flight vibration sensor at the current timestamp has generated abnormal vibration data, if it has been detected, determine the timestamp of the abnormal vibration data as the turbulence occurrence time; record the position of the positioning reference point as the initial target turbulence position, and record the height of the position of the positioning reference point as the height at which the turbulence occurs.

[0016] Preferably, as one possible implementation, the operation of calculating the EDR index based on QAR data includes:

[0017] S21. Calculate the standard deviation of vertical wind speed for the vertical wind component based on the QAR data. S22. Construct the aircraft's vertical acceleration response function; S23. Calculate the vertical acceleration response energy; S24. Invert the EDR value based on the vertical acceleration response energy.

[0018] Preferably, as one possible implementation: the standard deviation of the vertical wind speed of the vertical wind component is calculated based on the QAR data. The process involves: constructing the aircraft's vertical acceleration response function; calculating the vertical acceleration response energy; and retrieving the EDR value based on the vertical acceleration response energy, including the following steps:

[0019] The standard deviation of vertical wind speed for the vertical wind component is calculated based on the QAR data. ,include:

[0020] ;

[0021] in This represents the average vertical wind speed within the sliding time window. The standard deviation of vertical wind speed; The vertical wind component at the current timestamp;

[0022] Construct the aircraft's vertical acceleration response function:

[0023]

[0024] in, For turbulence frequency, The aircraft's inertial time constant is used; the standard deviation of acceleration obtained from real-time detection is corrected based on the aircraft's vertical acceleration response function. The corrected standard deviation of vertical acceleration was obtained and recorded as follows: ;

[0025] Then, the EDR value is retrieved based on the vertical acceleration response energy:

[0026] ;

[0027] in The standard deviation of vertical acceleration. For model response coefficient, This is airspeed.

[0028] Preferably, as one possible implementation, Ellrod_TI1 is used to characterize the intensity of mechanical turbulence; Ellrod_TI2 is used to characterize the deformation of the horizontal flow field.

[0029] Preferably, as one possible implementation, the bumpy area expanded in step S2, the bumpy area located by AMDAR data, and each of the bump indices in step S3 are compared and scored to output a scoring result, including the following steps:

[0030] S41. Define the bumpy event as a binary event: positive class is bumpy occurrence, negative class is no bumpy occurrence;

[0031] S42. Set thresholds for each turbulence index to generate ROC curves and calculate the four classification results:

[0032] YY: Index diagnosis indicates that the fluctuations actually occurred;

[0033] NY: The index did not diagnose the turbulence, but it actually occurred;

[0034] YN: The index indicates a fluctuation, but it did not actually occur;

[0035] NN: The index was not diagnosed and did not actually occur;

[0036] S42. Calculate PODY and PODN:

[0037] ;

[0038] The values ​​of PODY and PODN range from [0, 1]. The larger the value of PODY, the lower the false alarm rate; the larger the value of PODN, the lower the false alarm rate.

[0039] S43, with PODY as the ordinate, Plot the ROC curve for the x-axis;

[0040] S44. Calculate the area under the ROC curve (AUC) and select turbulence indices with AUC > 0.5 for fusion.

[0041] Preferably, as one possible implementation: a fused turbulence index is generated by weighted fusion of the turbulence index based on the scoring results, and turbulence regions are identified based on the fused turbulence index, including:

[0042] S51. Assign weights to the turbulence index based on AUC values: ;

[0043] S52. Perform a weighted summation on the normalized exponent values:

[0044] S53. Setting a fusion index threshold to identify bumpy areas includes: plotting the ROC curve of the fusion bump index based on historical data; determining the optimal threshold point on the ROC curve, which is the point on the ROC curve closest to the upper left corner; using the fusion index value corresponding to the optimal threshold point obtained in the previous step as the discrimination threshold; and determining a bumpy area when the real-time calculated fusion bump index is greater than or equal to the discrimination threshold.

[0045] Thirdly, this application provides a computing device, including a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement the aviation turbulence identification method based on multi-source data as described in the first aspect above.

[0046] Fourthly, this application provides a computer storage medium storing a computer program, which, when executed by a computer, implements an aviation turbulence identification method based on multi-source data as described in the first aspect.

[0047] The technical solution of this application has the following beneficial effects:

[0048] This application provides a method for identifying aviation turbulence based on multi-source data, including:

[0049] S1. Using ADS-B track data and aircraft in-flight reports, locate the location and altitude of the turbulence based on the flight number and turbulence time, and determine the initial target turbulence location;

[0050] S2. Calculate the EDR index based on QAR data, and perform a horizontal adjustment on the initial target bump position located in step S1. A vertical area of ​​600 meters is expanded to identify and output bumpy areas;

[0051] S3. Calculate at least several turbulence indices using gridded field data, including Brownian index, MOS CAT probability prediction factor, Dutton index, Ellrod_TI1, Ellrod_TI2, horizontal temperature gradient index, and CCAT index.

[0052] S4. Compare and score the bumpy area expanded in step S2, the bumpy area located by AMDAR data, and each of the bumpy indices in step S3, and output the scoring results. Based on the scoring results, the bumpy indices are weighted and fused to generate a fused bumpy index.

[0053] S5. Identify the bumpy area based on the fused bump index.

[0054] Analysis of the above technical solutions reveals that this invention addresses the ambiguity problem of positioning from a single data source (such as pilot reports alone) by combining ADS-B positioning (spatial accuracy) and QAR / EDR quantization (physical intensity) with hierarchical verification of grid field indices (environmental background). First, using ADS-B track data and aircraft in-flight reports, the location and altitude of turbulence are determined based on the flight number and turbulence time, establishing the initial target turbulence location, i.e., turbulence event localization (step 1). Then, area expansion (step 2) is implemented to multi-index fusion diagnosis (steps 3-5), constructing a closed-loop identification chain to accurately identify the final turbulence area. Attached Figure Description

[0055] To more clearly illustrate the technical solutions in the embodiments of this application or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of this application. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0056] Figure 1 A flowchart of an aviation turbulence identification method based on multi-source data provided in this application is shown;

[0057] Figure 2 This paper shows a schematic diagram of the structure of an aviation turbulence identification system based on multi-source data provided in this application;

[0058] Figure 3 A schematic diagram of the structure of a computing device provided in this application is shown. Detailed Implementation

[0059] To enable those skilled in the art to better understand the present application, the technical solutions in the embodiments of the present application will be clearly and completely described below with reference to the accompanying drawings.

[0060] In some of the processes described in the specification, claims, and accompanying drawings of this application, multiple operations appear in a specific order. However, it should be clearly understood that these operations may not be performed in the order they appear herein, or may be performed in parallel. The sequence numbers of the operations are merely used to distinguish different operations and do not themselves represent any execution order. Furthermore, these processes may include more or fewer operations, and these operations may be performed sequentially or in parallel. It should be noted that the terms "first," "second," etc., used herein are used to distinguish different messages, devices, modules, etc., and do not represent a sequential order, nor do they limit "first" and "second" to different types.

[0061] The technical solutions of the embodiments of this application will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of this application, and not all embodiments. Based on the embodiments of this application, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of this application.

[0062] Figure 1 A flowchart of an aviation turbulence identification method based on multi-source data is provided in this application embodiment, as follows: Figure 1 As shown, the method includes:

[0063] S1. Using ADS-B track data and aircraft in-flight reports, locate the location and altitude of the turbulence based on the flight number and turbulence time, and determine the initial target turbulence location;

[0064] S2. Calculate the EDR index based on QAR data, expand the initial target bumpy position located in step S1 by 1°×1° in the horizontal direction and 600 meters in the vertical direction, and identify and output the bumpy area.

[0065] Among them, QAR data and EDR index;

[0066] QAR data refers to high-frequency data recording systems installed on aircraft that continuously collect hundreds of flight parameters (such as acceleration, attitude, airspeed, engine status, etc.). QAR data is primarily used for airline maintenance monitoring and flight quality analysis. Unlike FDR (black box) data, QAR data can be extracted quickly and periodically without requiring an accident to trigger it.

[0067] The aforementioned EDR (Eddy Dissipation Rate) is the rate of energy dissipation in turbulence, a physical quantity characterizing the intensity of atmospheric turbulence, and a turbulence metric standardized by the International Civil Aviation Organization (ICAO).

[0068] EDR Index Calculation Formula (Core Principles):

[0069] ;

[0070] : Vertical wind speed standard deviation (reflects the intensity of turbulent fluctuations); Aircraft-specific constants (related to aircraft inertia and sensor response); Vacuum speed; the EDR index can objectively quantify the intensity of turbulence and is not affected by the pilot's subjective feelings.

[0071] S3. Calculate at least several turbulence indices using gridded field data, including Brownian index, MOS CAT probability prediction factor, Dutton index, Ellrod_TI1, Ellrod_TI2, horizontal temperature gradient index, and CCAT index.

[0072] The gridded field data described above divides the Earth's atmosphere into a three-dimensional grid (e.g., 0.25° × 0.25° longitude / latitude, 37 vertical layers), with each grid point storing historical meteorological element values. Typical elements of the gridded field data include: temperature, air pressure, and humidity (directly affecting thermal turbulence); wind speed and direction (determining dynamic turbulence); vorticity and vertical shear (key driving factors of turbulence).

[0073] Ellrod_TI1 (vertical wind shear-dominated type) is used to characterize the intensity of mechanical turbulence; Ellrod_TI2 (horizontal deformation-dominated type) is used to characterize the deformation of the horizontal flow field.

[0074] S4. Compare and score the bumpy area expanded in step S2, the bumpy area located by AMDAR data, and each of the bumpy indices in step S3, and output the scoring results. Based on the scoring results, the bumpy indices are weighted and fused to generate a fused bumpy index.

[0075] S5. Identify the bumpy area based on the fused bump index.

[0076] Analysis of the above technical solutions reveals that this invention addresses the ambiguity problem of positioning from a single data source (such as pilot reports alone) by combining ADS-B positioning (spatial accuracy) and QAR / EDR quantization (physical intensity) with hierarchical verification of grid field indices (environmental background). First, using ADS-B track data and aircraft in-flight reports, the location and altitude of turbulence are determined based on the flight number and turbulence time, establishing the initial target turbulence location, i.e., turbulence event localization (step 1). Then, area expansion (step 2) is implemented to multi-index fusion diagnosis (steps 3-5), constructing a closed-loop identification chain to accurately identify the final turbulence area.

[0077] Preferably, as one possible implementation, the operation of locating the bumpy position in step S1 includes:

[0078] The operation of locating the bump location in step S1 includes:

[0079] S11. Obtain the flight number, and parse the timestamp, longitude, latitude, and altitude fields of the corresponding ADS-B waypoint based on the flight number; S12. Match the turbulence occurrence time in the aircraft's in-flight report based on the ADS-B waypoint, and determine the initial target turbulence location.

[0080] Preferably, as one possible implementation; step S12, matching the turbulence occurrence time in the aircraft's in-flight report and determining the initial target turbulence location, includes:

[0081] S121. Align the timestamps in the aircraft's in-flight reports with the data timeline of the ADS-B trackpoints;

[0082] S122. Select ADS-B waypoints within ±30 seconds before and after the report time;

[0083] S123. If there are multiple ADS-B track points, select the geometric center coordinates as the positioning reference point; if there is only one ADS-B track point, determine that track point as the positioning reference point, determine whether the in-flight vibration sensor at the current timestamp has generated abnormal vibration data, if it has been detected, determine the timestamp of the abnormal vibration data as the turbulence occurrence time; record the position of the positioning reference point as the initial target turbulence position, and record the height of the position of the positioning reference point as the height at which the turbulence occurs.

[0084] Preferably, as one possible implementation, the operation of calculating the EDR index based on QAR data includes:

[0085] S21. Calculate the standard deviation of vertical wind speed for the vertical wind component based on the QAR data. S22. Construct the aircraft's vertical acceleration response function; S23. Calculate the vertical acceleration response energy; S24. Invert the EDR value based on the vertical acceleration response energy.

[0086] Preferably, as one possible implementation: the standard deviation of the vertical wind speed of the vertical wind component is calculated based on the QAR data. The process involves: constructing the aircraft's vertical acceleration response function; calculating the vertical acceleration response energy; and retrieving the EDR value based on the vertical acceleration response energy, including the following steps:

[0087] The standard deviation of vertical wind speed for the vertical wind component is calculated based on the QAR data. ,include:

[0088] ;

[0089] in This represents the average vertical wind speed within the sliding time window. The standard deviation of vertical wind speed; The vertical wind component at the current timestamp;

[0090] Construct the aircraft's vertical acceleration response function:

[0091] ;

[0092] Among them, For turbulence frequency, The aircraft's inertial time constant is used; the standard deviation of acceleration obtained from real-time detection is corrected based on the aircraft's vertical acceleration response function. The corrected standard deviation of vertical acceleration was obtained and recorded as follows: ;

[0093] Then, the EDR value is retrieved based on the vertical acceleration response energy:

[0094] ;

[0095] in The standard deviation of vertical acceleration. For model response coefficient, This is airspeed.

[0096] Preferably, as one possible implementation, the bumpy area expanded in step S2, the bumpy area located by AMDAR data, and each of the bump indices in step S3 are compared and scored to output a scoring result, including the following steps:

[0097] S41. Define the bumpy event as a binary event: positive class is bumpy occurrence, negative class is no bumpy occurrence;

[0098] S42. Set thresholds for each turbulence index to generate ROC curves and calculate the four classification results:

[0099] YY: Index diagnosis indicates that the fluctuations actually occurred;

[0100] NY: The index did not diagnose the turbulence, but it actually occurred;

[0101] YN: The index indicates a fluctuation, but it did not actually occur;

[0102] NN: The index was not diagnosed and did not actually occur;

[0103] S42. Calculate PODY and PODN:

[0104] ;

[0105] The values ​​of PODY and PODN range from [0, 1]. The larger the value of PODY, the lower the false alarm rate; the larger the value of PODN, the lower the false alarm rate.

[0106] S43, with PODY as the ordinate, Plot the ROC curve for the x-axis;

[0107] S44. Calculate the area under the ROC curve (AUC) and select turbulence indices with AUC > 0.5 for fusion.

[0108] Preferably, as one possible implementation: a fused turbulence index is generated by weighted fusion of the turbulence index based on the scoring results, and turbulence regions are identified based on the fused turbulence index, including:

[0109] S51. Assign weights to the turbulence index based on AUC values: S52. Perform a weighted summation of the normalized exponent values:

[0110] S53. Setting a fusion index threshold to identify bumpy areas includes: plotting the ROC curve of the fusion bump index based on historical data; determining the optimal threshold point on the ROC curve, which is the point on the ROC curve closest to the upper left corner; using the fusion index value corresponding to the optimal threshold point obtained in the previous step as the discrimination threshold; and determining a bumpy area when the real-time calculated fusion bump index is greater than or equal to the discrimination threshold.

[0111] The embodiments of the present invention employ one of the following two methods:

[0112] S51', Assign weights to the turbulence index based on AUC values: ;

[0113] S52', Perform a weighted summation of the normalized exponent values:

[0114] S53' Setting a fusion index threshold to identify bumpy areas includes: a) Obtaining the distribution of the fusion bump index in historical data when bumpy events occur and when they do not; b) Calculating PODY (hit rate) and PODN (accuracy without bumps) under different candidate thresholds; c) Selecting the candidate threshold that maximizes (PODY + PODN) as the fusion index threshold; d) For the fusion bump index calculated in real time, when the index value is greater than or equal to the threshold, it is determined to be a bumpy area.

[0115] However, this approach involves many steps and may be overly detailed. A simpler method is to directly use the ROC curve to determine the threshold, as the ROC curve was already introduced in previous steps and can be used here:

[0116] The embodiments of the present invention employ one of the following two methods:

[0117] S51'', Assigning weights to the turbulence index based on AUC values: ;

[0118] S52'', Perform a weighted summation of the normalized exponent values:

[0119] S53'', Setting a fusion index threshold to identify bumpy areas includes: a) Plotting the ROC curve of the fusion bumpy index based on historical data; b) Determining the optimal threshold point on the ROC curve, which is the point closest to the upper left corner of the ROC curve; c) Using the fusion index value corresponding to the optimal threshold point obtained in step b) as the discrimination threshold; d) When the real-time calculated fusion bumpy index is greater than or equal to the discrimination threshold, it is determined to be a bumpy area. However, since the ROC curve has already been plotted for the individual index in the previous steps, it is reasonable to plot the ROC curve for the fusion index again, because the fusion index is a new index.

[0120] Here is a specific example: Automatic Dependent Surveillance-Broadcast (ADS-B) has been identified by the International Civil Aviation Organization (ICAO) as the primary future air traffic control technology. Aircraft equipped with ADS-B transmitting devices automatically broadcast flight identification marks, positions, and flight status information to other aircraft equipped with ADS-B receiving devices and ground receiving stations via air traffic data links. A flight's ADS-B flight path consists of multiple trackpoints, each carrying multiple fields of information (timestamp, longitude, latitude, altitude, heading, speed, etc.). In this embodiment of the invention, ADS-B track data and aircraft in-flight reports are used to pinpoint the location and altitude of turbulence. Based on ADS-B track data, combined with the flight number and the time of turbulence, this embodiment of the invention locates the aircraft's position at high altitude where turbulence occurred, identifying the turbulence area. This positioning method is more objective and more accurate than the position reported by the crew via voice communication.

[0121] The specific evaluation methods are as follows:

[0122] Based on the turbulence areas located by aircraft air reports, ADS-B data, etc., the occurrence of turbulence is considered a positive event, and the absence of turbulence is considered a negative event. Therefore, aircraft turbulence events are considered as binary events of occurrence and non-occurrence. At the same time, the turbulence index diagnosis results and the actual aircraft observation results (turbulence located by aircraft air reports, AMDAR data, etc.) can be divided into 4 cases, as shown in Table 1.

[0123] Table 1. Classification of Aircraft Observation and Index Diagnosis

[0124]

[0125] Based on the above four classifications, PODY (diagnostic accuracy with turbulence), PODN (diagnostic accuracy without turbulence), and the area under the curve (AUC) of the relative action characteristic curve (ROC) are selected as evaluation indicators for the diagnostic effectiveness of the turbulence index.

[0126] The diagnostic accuracy (PODY) for turbulence is expressed as:

[0127]

[0128] The diagnostic accuracy (PODN) for bump-free conditions is expressed as:

[0129]

[0130] The values ​​of PODY and PODN range from [0, 1]. The larger the value of PODY, the lower the false alarm rate, and the larger the value of PODN, the lower the false alarm rate.

[0131] For binary classification events, PODY and 1-PODN together form the vertical and horizontal axes of the ROC plot. The ROC curve is generated by applying a series of thresholds. Therefore, the ROC curve reflects a comprehensive indicator of overall performance. The area near the top left corner of the ROC plot (i.e., high hit rate and low false positive rate) indicates positive skill, while the diagonal represents "no skill". To quantify skill level, we calculate the area integral (AUC) of the ROC curve on the horizontal axis, ranging from [0, 1]. A larger AUC indicates better overall diagnostic ability.

[0132] The specific steps of the fusion turbulence identification algorithm are as follows: 1) Calculate seven turbulence indices using grid analysis data: Brownian index, MOSCAT probability prediction factor, Dutton index, Ellrod_TI1, Ellrod_TI2, horizontal temperature gradient index, and CCAT index; 2) Set different thresholds for each index and calculate the corresponding PODY and PODN for each threshold; 3) Plot the ROC curve based on the correspondence between PODY and 1-PODN, and calculate the area under the curve (AUC). When the AUC area exceeds 0.5, the turbulence index is considered to have positive skill; 4) Select the turbulence indices with positive skill and perform weighted average fusion to obtain the fused turbulence index, thus identifying turbulent areas.

[0133] Example 2

[0134] Figure 2 This application provides a schematic diagram of the structure of an aviation turbulence identification system based on multi-source data, as shown in the embodiment of the present application. Figure 2 As shown, the system includes:

[0135] The initial positioning module 10 is used to locate the location and altitude of the turbulence based on the flight number and turbulence time using ADS-B track data and aircraft in-flight reports, and to determine the initial target turbulence location.

[0136] The secondary positioning module 20 is used to calculate the EDR index based on QAR data and to perform horizontal positioning on the initial target bump position located in step S1. A vertical area of ​​600 meters is expanded to identify and output bumpy areas;

[0137] The calculation and acquisition module 30 is used to calculate at least a variety of turbulence indices, including Brownian index, MOS CAT probability prediction factor, Dutton index, Ellrod_TI1, Ellrod_TI2, horizontal temperature gradient index, and CCAT index, using grid field data.

[0138] The fusion module 40 is used to compare and score the extended bump area, the bump area located by AMDAR data, and each of the bump indices in step S3, and output a score result. Based on the score result, the bump indices are weighted and fused to generate a fused bump index.

[0139] The identification processing module 50 is used to identify bumpy areas based on the fused bump index.

[0140] Figure 2 The aforementioned multi-source data-based aircraft turbulence identification system can perform... Figure 1 The implementation principle and technical effects of the aviation turbulence identification method based on multi-source data described in the illustrated embodiment will not be repeated here. The specific methods by which each module and unit of the aviation turbulence identification system based on multi-source data performs its operations have been described in detail in the embodiments related to this method, and will not be elaborated upon here.

[0141] Example 3

[0142] In one possible design, Figure 2 The aviation turbulence identification system based on multi-source data in the illustrated embodiment can be implemented as a computing device, such as... Figure 3 As shown, the computing device may include a storage component 31 and a processing component 32;

[0143] The storage component 31 stores one or more computer instructions, wherein the one or more computer instructions are invoked and executed by the processing component 32.

[0144] The processing component 32 is used for the above Figure 1 The embodiment describes an aviation turbulence identification method based on multi-source data.

[0145] The processing component 32 may include one or more processors to execute computer instructions to complete all or part of the steps in the above-described method. Alternatively, the processing component may be implemented as one or more application-specific integrated circuits (ASICs), digital signal processors (DSPs), digital signal processing devices (DSPDs), programmable logic devices (PLDs), field-programmable gate arrays (FPGAs), controllers, microcontrollers, microprocessors, or other electronic components to perform the above-described method.

[0146] Storage component 31 is configured to store various types of data to support operations at the terminal. The storage component can be implemented by any type of volatile or non-volatile storage device or a combination thereof, such as static random access memory (SRAM), electrically erasable programmable read-only memory (EEPROM), erasable programmable read-only memory (EPROM), programmable read-only memory (PROM), read-only memory (ROM), magnetic storage, flash memory, magnetic disk, or optical disk.

[0147] Of course, computing devices may also include other components, such as input / output interfaces, display components, communication components, etc.

[0148] Input / output interfaces provide interfaces between processing components and peripheral interface modules, which can be output devices, input devices, etc.

[0149] The communication components are configured to facilitate wired or wireless communication between computing devices and other devices.

[0150] The computing device can be a physical device or an elastic computing host provided by a cloud computing platform. In this case, the computing device can refer to a cloud server, and the aforementioned processing components, storage components, etc., can be basic server resources rented or purchased from the cloud computing platform.

[0151] Example 4

[0152] This application also provides a computer storage medium storing a computer program, which, when executed by a computer, can perform the above-described functions. Figure 1 The embodiment shown is an aviation turbulence identification method based on multi-source data.

[0153] Those skilled in the art will clearly understand that, for the sake of convenience and brevity, the specific working processes of the systems, devices, and units described above can be referred to the corresponding processes in the foregoing method embodiments, and will not be repeated here.

[0154] The device embodiments described above are merely illustrative. The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; that is, they may be located in one place or distributed across multiple network units. Some or all of the modules can be selected to achieve the purpose of this embodiment according to actual needs. Those skilled in the art can understand and implement this without any creative effort.

[0155] Through the above description of the embodiments, those skilled in the art can clearly understand that each embodiment can be implemented by means of software plus necessary general-purpose hardware platforms, and of course, it can also be implemented by hardware. Based on this understanding, the above technical solutions, in essence or the part that contributes to the prior art, can be embodied in the form of a software product. This computer software product can be stored in a computer-readable storage medium, such as ROM / RAM, magnetic disk, optical disk, etc., and includes several instructions to cause a computer device (which may be a personal computer, server, etc.) to execute the methods described in the various embodiments or some parts of the embodiments.

[0156] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of this application, and are not intended to limit them. Although this application has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of this application.

Claims

1. A method for identifying aircraft turbulence based on multi-source data, characterized in that, include: S1. Using ADS-B track data and aircraft in-flight reports, locate the location and altitude of the turbulence based on the flight number and turbulence time, and determine the initial target turbulence location; S2. Calculate the EDR index based on QAR data, expand the initial target bumpy position located in step S1 by 1°×1° in the horizontal direction and 600 meters in the vertical direction, and identify and output the bumpy area. S3. Calculate at least several turbulence indices using gridded field data, including the Brownian index, MOS CAT probability prediction factor, Dutton index, Ellrod_TI1, Ellrod_TI2, horizontal temperature gradient index, and CCAT index; Ellrod_TI1 is used to characterize the intensity of mechanical turbulence; Ellrod_TI2 is used to characterize the deformation of the horizontal flow field; S4. Compare and score the bumpy area expanded in step S2, the bumpy area located by AMDAR data, and each of the bumpy indices in step S3, and output the scoring results. Based on the scoring results, the bumpy indices are weighted and fused to generate a fused bumpy index. S5. Identify the bumpy area based on the fused bump index; The process of comparing and scoring the expanded bumpy area in step S2, the bumpy area located by AMDAR data, and the various bump indices in step S3 to output a scoring result includes the following steps: S41. Define the bumpy event as a binary event: positive class is bumpy occurrence, negative class is no bumpy occurrence; S42. Set thresholds for each turbulence index to generate ROC curves and calculate the four classification results: YY: Index diagnosis indicates that the fluctuations actually occurred; NY: The index did not diagnose the turbulence, but it actually occurred; YN: The index indicates a fluctuation, but it did not actually occur; NN: The index was not diagnosed and did not actually occur; S42. Calculate PODY and PODN: ； The values ​​of PODY and PODN range from [0, 1]. The larger the value of PODY, the lower the false alarm rate; the larger the value of PODN, the lower the false alarm rate. S43, with PODY as the ordinate, Plot the ROC curve for the x-axis; S44. Calculate the area under the ROC curve (AUC) and select turbulence indices with AUC > 0.5 for fusion.

2. The method according to claim 1, characterized in that, The operation of locating the bump location in step S1 includes: S11. Obtain the flight number, and parse the timestamp, longitude, latitude, and altitude fields of the corresponding ADS-B waypoint based on the flight number; S12. Match the turbulence occurrence time in the aircraft's in-flight report based on the ADS-B waypoint, and determine the initial target turbulence location.

3. The method according to claim 2, characterized in that, Step S12, which involves matching the turbulence occurrence time in the aircraft's in-flight report and determining the initial target turbulence location, includes: S121. Align the timestamps in the aircraft's in-flight reports with the data timeline of the ADS-B trackpoints; S122. Select ADS-B waypoints within ±30 seconds before and after the report time; S123. If there are multiple ADS-B track points, select the geometric center coordinates as the positioning reference point; if there is only one ADS-B track point, determine that track point as the positioning reference point, determine whether the in-flight vibration sensor at the current timestamp has generated abnormal vibration data, if it has been detected, determine the timestamp of the abnormal vibration data as the turbulence occurrence time; record the position of the positioning reference point as the initial target turbulence position, and record the height of the position of the positioning reference point as the height at which the turbulence occurs.

4. The method according to claim 1, characterized in that, The operation of calculating the EDR index based on QAR data includes: S21. Calculate the standard deviation of vertical wind speed for the vertical wind component based on the QAR data. S22. Construct the aircraft's vertical acceleration response function; S23. Calculate the vertical acceleration response energy; S24. Invert the EDR value based on the vertical acceleration response energy.

5. The method according to claim 4, characterized in that, S21. Calculate the standard deviation of vertical wind speed for the vertical wind component based on the QAR data. S22. Construct the aircraft's vertical acceleration response function; S23. Calculate the vertical acceleration response energy; S24. Invert the EDR value based on the vertical acceleration response energy, including the following steps: The standard deviation of vertical wind speed for the vertical wind component is calculated based on the QAR data. ,include: ； in This represents the average vertical wind speed within the sliding time window. The standard deviation of vertical wind speed; The vertical wind component at the current timestamp; Construct the aircraft's vertical acceleration response function: ;in, For turbulence frequency, The aircraft's inertial time constant is used; the standard deviation of acceleration obtained from real-time detection is corrected based on the aircraft's vertical acceleration response function. The corrected standard deviation of vertical acceleration was obtained and recorded as follows: ; Then, the EDR value is retrieved based on the vertical acceleration response energy: ； in The standard deviation of vertical acceleration. For model response coefficient, This is airspeed.

6. The method according to claim 5, characterized in that, Based on the scoring results, a weighted blizzard index is generated by fusing the blizzard indices. Bumpy regions are then identified based on the fusing blizzard index, including: S51. Assign weights to the turbulence index based on AUC values: ; S52. Perform a weighted summation on the normalized exponent values: S53. Setting a fusion index threshold to identify bumpy areas includes: plotting the ROC curve of the fusion bump index based on historical data; determining the optimal threshold point on the ROC curve, which is the point on the ROC curve closest to the upper left corner; using the fusion index value corresponding to the optimal threshold point obtained in the previous step as the discrimination threshold; and determining a bumpy area when the real-time calculated fusion bump index is greater than or equal to the discrimination threshold.

7. A computing device, characterized in that, It includes a processing component and a storage component; the storage component stores one or more computer instructions; the one or more computer instructions are invoked and executed by the processing component to implement the aviation turbulence identification method based on multi-source data as described in any one of claims 1 to 6.

8. A computer storage medium, characterized in that, The device contains a computer program that, when executed by a computer, implements an aviation turbulence identification method based on multi-source data as described in any one of claims 1 to 6.

Images