Method and device for determining aircraft landing turbulence index, storage medium and equipment

Abstract

The application provides a method and device for determining an aircraft turbulence index, a storage medium and equipment, and belongs to the field of aviation safety in civil aviation. The method is as follows: collecting QAR data of all flights of the same aircraft model at low altitude and near ground level; calculating the mean value of the pilot's operation amplitude and the wind field related variable of each flight per second based on the QAR data; searching for matching data in the flight data of all flights using the K nearest neighbor method based on the wind field related variable for each second of data before the aircraft lands, forming a data set, and calculating a quantitative index of the pilot's operation amplitude fluctuation degree based on the data set; fitting the quantitative index of the pilot's operation amplitude fluctuation degree based on the wind field related variable to construct a semi-parametric regression model; and obtaining a turbulence index calculation formula based on the semi-parametric regression model. The application improves the accuracy of the turbulence index and can more accurately reflect the quantitative index of the influence of turbulence on the aircraft.

Description

TECHNICAL FIELD

[0001] The present application belongs to the field of aviation safety in civil aviation industry, and particularly relates to a method and device for determining a turbulence index of an aircraft landing, a storage medium and equipment. BACKGROUND

[0002] The turbulence index can effectively reflect the change range of pilot operation and the turbulence degree of airflow, and can stably capture the influence of airflow on flight operation. Specifically, the more turbulent the airflow is, the larger the index value is, and the larger the IQR value of pilot operation is. The index can comprehensively consider the influence of various meteorological factors and geographical environment on turbulence, and the index is more dispersed and has a larger value in seasons or airport locations with complex airflow changes.

[0003] Hon K K et al. applied an XGBoost algorithm to combine a series of traditional "turbulence indices" generated by a numerical weather prediction (NWP) model to generate effective aviation turbulence prediction through optimal combination, so as to improve the prediction ability of aviation turbulence events.

[0004] Zhang Q et al. proposed a turbulence target detection method based on BP (Back Propagation) neural network multi-class classification, which can effectively detect the strength of turbulence or judge whether turbulence exists or not by learning a large amount of airborne weather radar data and using the classification function of neural network to establish the relationship between radar echo and turbulence intensity.

[0005] Cai J et al. used the observation data of the Terminal Doppler Weather Radar (TDWR) of the Hong Kong Observatory to propose an artificial neural network-based method for predicting and classifying the cubic root of turbulence dissipation rate, which performs well especially in detecting moderate and above turbulence.

[0006] Huang R et al. combined two traditional methods for estimating Eddy Dissipation Rate (EDR) to propose a new EDR estimation program based on QAR data, which can accurately reflect the strength of air turbulence.

[0007] Mizuno S et al. proposed a method combining Principal Component Analysis (PCA), K-Means clustering and Support Vector Classification (SVC) to predict turbulence risk using meteorological data and QAR data, which significantly improves the prediction performance of the model through feature dimension reduction and clustering analysis.

[0008] Currently, there are two methods to determine the turbulence index. One is that the airport uses meteorological detection equipment to detect the air flow intensity and changes near the airport through real-time wind field information in a large range, so as to determine the turbulence intensity and issue a warning. However, the meteorological radar beam is wide and the spatial resolution is low, which makes it difficult to accurately capture the subtle changes of wind speed and direction in the local area. The other is a method of calculating turbulence index based on QAR data. However, most of the current methods for evaluating wind shear and turbulence mainly use unsupervised learning methods, which lack direct correlation with actual flight operation data, resulting in limited accuracy and reliability of the evaluation results in actual application. SUMMARY

[0009] The purpose of the present application is to solve the problems existing in the prior art, and to provide a method, device, storage medium and equipment for determining the turbulence index of an aircraft landing, which improves the accuracy of the turbulence index and can more accurately reflect the quantitative index of the influence of turbulence on the aircraft.

[0010] The present application is achieved by the following technical solutions:

[0011] In a first aspect, the present application provides a method for determining the turbulence index of an aircraft landing, which specifically comprises the following steps:

[0012] Step 1: Collecting QAR data of all flights of the same aircraft model at low altitude and near ground stage in the same airport;

[0013] Step 2: Calculating the pilot operation amplitude mean value and wind field related variables of each flight per second based on the QAR data;

[0014] Step 3: Based on the wind field related variables, using K-nearest neighbor method to search for matching data in the flight data of all flights, forming a data set, and calculating the quantitative index of pilot operation amplitude fluctuation degree based on the data set;

[0015] Step 4: Fitting the quantitative index of pilot operation amplitude fluctuation degree based on the wind field related variables, and constructing a semi-parametric regression model;

[0016] Step 5: Obtaining the turbulence index calculation formula based on the semi-parametric regression model.

[0017] Further improvements of the present application are as follows:

[0018] In step 1, the QAR data includes: height, flight stage, heading, wind direction, wind speed, automatic pilot state, captain pitch operation amplitude, co-pilot pitch operation amplitude, captain roll operation amplitude, and co-pilot roll operation amplitude.

[0019] Further improvements of the present application are as follows:

[0020] In the step 2,

[0021] The average of the pilot's pitch operation amplitude = the sum of the pitch operation amplitudes of the pilot collected in each second / the total number of the pitch operation amplitudes of the pilot in each second.

[0022] The average of the pilot's roll operation amplitude = the sum of the roll operation amplitudes of the pilot collected in each second / the total number of the roll operation amplitudes of the pilot in each second.

[0023] The further improvement of the present application is that:

[0024] In the step 2, the wind field related variables include the angle between the wind direction and the heading, the wind speed change value, the wind direction change value, the wind speed standard deviation in the past four seconds and the wind direction fluctuation in the past four seconds.

[0025] The further improvement of the present application is that:

[0026] In the step 3, the quantitative index of the pilot operation amplitude fluctuation degree is calculated based on the data set, and the operation includes:

[0027] The 10% quantile and the 90% quantile of the average of the pilot's pitch operation amplitude in the data set are calculated, and the interquartile range IQR is obtained by the difference between the two, as the quantitative index of the pilot's pitch operation amplitude fluctuation degree corresponding to each second;

[0028] The 10% quantile and the 90% quantile of the average of the pilot's roll operation amplitude in the data set are calculated, and the interquartile range IQR is obtained by the difference between the two, as the quantitative index of the pilot's roll operation amplitude fluctuation degree corresponding to each second;

[0029] The quantitative index of the pilot's pitch operation amplitude fluctuation degree corresponding to each second or the quantitative index of the pilot's roll operation amplitude fluctuation degree corresponding to each second is calculated based on each data set in turn.

[0030] The further improvement of the present application is that:

[0031] The operation of the step 4 includes:

[0032] The height is taken as the nonlinear part, and the wind speed, the angle between the wind direction and the heading, the wind speed change value, the wind direction change value, the wind speed standard deviation in the past four seconds and the wind direction fluctuation in the past four seconds are taken as the linear part, the quantitative index of the pilot's pitch operation amplitude fluctuation degree is fitted, and a semi-parametric regression model of the pitch operation is constructed;

[0033] The quantification index of the roll operation amplitude fluctuation degree of the pilot is fitted by taking the height as a nonlinear part, and taking the wind speed, the angle between the wind direction and the heading, the wind speed change value, the wind direction change value, the wind speed standard deviation in the past four seconds and the wind direction fluctuation in the past four seconds as linear parts, and a semi-parametric regression model of the roll operation is constructed.

[0034] Further improvements of the present application are:

[0035] The operation in the step 5 comprises:

[0036] The influence of the height on the semi-parametric regression model of the pitch operation is removed, and a turbulence index calculation formula for the pitch operation of the pilot is obtained.

[0037] The influence of the height on the semi-parametric regression model of the roll operation is removed, and a turbulence index calculation formula for the roll operation of the pilot is obtained.

[0038] In a second aspect, the present application provides a determination device of the landing turbulence index of an airplane, which is used to execute the steps in the determination method of the landing turbulence index of the airplane, and the system comprises:

[0039] A data acquisition unit is configured to acquire QAR data of all airplanes of the same type of airplanes of the same airport in a low-altitude and near-ground stage.

[0040] A first calculation unit is configured to calculate the pilot operation amplitude mean value and the wind field related variable of each airplane of each flight per second based on the QAR data.

[0041] A second calculation unit is configured to search for matched data in the flight data of all airplanes of all flights based on the wind field related variable by using a K-neighbor method, form a data set, and calculate the quantification index of the pilot operation amplitude fluctuation degree based on the data set.

[0042] A model construction unit is configured to fit the quantification index of the pilot operation amplitude fluctuation degree based on the wind field related variable, and construct a semi-parametric regression model.

[0043] A turbulence index determination unit is configured to obtain a turbulence index calculation formula based on the semi-parametric regression model.

[0044] In a third aspect, the present application provides a computer readable storage medium, which stores at least one computer executable program, and the at least one program is executed by the computer to make the computer execute the steps in the determination method of the landing turbulence index of the airplane.

[0045] In a fourth aspect, the present application provides a computer device comprising a memory and a processor, wherein the memory stores a computer program, and wherein the processor implements the steps of the method for determining the turbulence index of an airplane landing when executing the computer program.

[0046] Compared with the prior art, the present application has the following advantages:

[0047] The present application is based on the specific rod operation fluctuation of the pilot to guide the turbulence index evaluation, and the 10%-90% distance of the pilot's operation behavior under specific airflow conditions is used as the basis to calculate the quantitative index of the pilot's operation amplitude fluctuation degree. A semi-parametric regression model is obtained by fitting the quantitative index of the pilot's operation amplitude fluctuation degree, wherein the height variable is a non-parametric part, and the remaining variables are parameter parts. The parameter part of the semi-parametric regression model is used as a new turbulence index after removing the height influence, which improves the accuracy of the turbulence index and can more accurately reflect the quantitative index of the influence of turbulence on the airplane.

[0048] For a new airplane, through the technical index of the present application, the influence degree of turbulence on the airplane roll operation or pitch operation at each second can be calculated and predicted under the condition of obtaining the specific value of QAR data, and the higher the index, the greater the influence degree. BRIEF DESCRIPTION OF DRAWINGS

[0049] Figure 1 The flowchart of the method for determining the turbulence index of an airplane landing in an embodiment of the present application is shown in

[0050] Figure 2 The turbulence index of a flight with relatively stable airflow is shown in

[0051] Figure 3 The turbulence index of a flight with relatively turbulent airflow is shown in DETAILED DESCRIPTION

[0052] The present application will be further described in detail below with reference to the accompanying drawings:

[0053] During the flight, the pilot will take appropriate operation measures according to the actual wind field, such as adjusting the flight attitude, speed and height, etc. These operation data can directly reflect the pilot's perception and response ability to turbulence events, and are an important basis for evaluating the influence of turbulence events on flight safety.

[0054] Based on this, the present application provides a method for determining the turbulence index of an airplane landing, as shown in Figure 1 The method comprises the following steps:

[0055] Step 1: Collect the QAR data of all flights of the same type of airplane at low altitude and near ground stage in the same airport. Step 2: Calculate the turbulence index of the airplane landing according to the QAR data.

[0056] Step 2, calculate the pilot operation amplitude mean value and the wind field related variable of each flight aircraft per second based on the QAR data respectively;

[0057] Step 3, based on the wind field related variable, search the matched data in all flight aircraft flight data by using K-Nearest Neighbors (KNN) method for each second of the aircraft landing, form a data set, and calculate the quantification index of the pilot operation amplitude fluctuation degree based on the data set;

[0058] Step 4, fit the quantification index of the pilot operation amplitude fluctuation degree based on the wind field related variable, and construct a semi-parametric regression model;

[0059] Step 5, obtain the turbulence index calculation formula based on the semi-parametric regression model.

[0060] Further, in the step 1,

[0061] The QAR data includes: altitude_ALTITUDE, flight phase_FLIGHT_PHASE, heading_HEADING, wind direction_WINDIR, wind speed_WIN_SPD, autopilot status_AP_STATUS, captain pitch operation amplitude PITCH_CAPT, first officer pitch operation amplitude PITCH_FO, captain roll operation amplitude ROLL_CAPT, and first officer roll operation amplitude ROLL_FO.

[0062] The collected QAR data is used as the data basis of the whole model.

[0063] Further, in the step 2,

[0064] Generally, the number N of the collected pilot (captain or first officer) pitch operation amplitude or roll operation amplitude per second is greater than 1, in order to facilitate calculation and improve calculation accuracy, the pilot operation amplitude mean value of each flight aircraft per second in the present application includes the pilot pitch operation amplitude mean value and the pilot roll operation amplitude mean value.

[0065] The pilot pitch operation amplitude mean value = the sum of the collected captain or first officer pitch operation amplitude per second / the total number of the collected captain or first officer pitch operation amplitude per second;

[0066] For the calculation of the pilot pitch operation amplitude mean value, if the captain pitch operation amplitude has a non-zero value in the second, that is, the captain has generated an operation, at this time, the captain pitch operation amplitude data is used for calculation, then

[0067] The average of the pilot's pitch operation amplitude = the sum of all the data of the captain's pitch operation amplitude PITCH_CAPT in the second / the total number of the captain's pitch operation amplitude in the second;

[0068] If the captain's pitch operation amplitude is all zero, that is, the captain does not operate, at this time, the copilot's pitch operation amplitude data is used for calculation, then

[0069] The average of the pilot's pitch operation amplitude = the sum of all the data of the copilot's pitch operation amplitude PITCH_FO in the second / the total number of the copilot's pitch operation amplitude in the second.

[0070] The average of the pilot's roll operation amplitude = the sum of the captain's or copilot's roll operation amplitude collected in each second / the total number of the captain's or copilot's roll operation amplitude in each second.

[0071] For the calculation of the average of the pilot's roll operation amplitude, if the captain's roll operation amplitude has a non-zero value in the second, that is, the captain operates, at this time, the captain's roll operation amplitude data is used for calculation, then

[0072] The average of the pilot's roll operation amplitude = the sum of all the data of the captain's roll operation amplitude ROLL_CAPT in the second / the total number of the captain's roll operation amplitude in the second;

[0073] If the captain's roll operation amplitude is all zero, that is, the captain does not operate, at this time, the copilot's roll operation amplitude data is used for calculation, then

[0074] The average of the pilot's roll operation amplitude = the sum of all the data of the copilot's roll operation amplitude ROLL_FO in the second / the total number of the copilot's roll operation amplitude in the second.

[0075] Further, in step 2,

[0076] The wind field related variables include the wind direction and heading angle, the wind speed change value, the wind direction change value, the wind speed standard deviation in the past four seconds, and the wind direction fluctuation in the past four seconds, and the specific operation includes:

[0077] The wind direction and heading angle WINHEAD = wind direction_WINDIR - heading_HEADING, and the calculated difference is unified to the interval [-180, 180), and further, the specific operation includes:

[0078] First, the difference between the wind direction_WINDIR and the heading_HEADING, that is, _WINDIR-_HEADING, is calculated, and the wind direction_WINDIR and the heading_HEADING data are originally in the interval [0, 360);

[0079] If the difference is greater than or equal to 180 degrees, it means that the included angle is out of the range of [-180, 180), and the difference between the wind direction _WINDIR and the heading _HEADING is adjusted to the interval [-180, 180) by subtracting 360 degrees, that is, _WINDIR-_HEADING-360; similarly, if the difference is less than -180 degrees, the difference between the wind direction _WINDIR and the heading _HEADING is adjusted to the target interval [-180, 180) by adding 360 degrees, that is, _WINDIR-_HEADING+360; if the difference is already in the range of [-180, 180), the difference is directly used as the included angle.

[0080] The wind speed change value WINSPD_DIFF is the wind speed _WIN_DIR of the current second minus the wind speed _WIN_DIR of the last second.

[0081] The wind direction change value WINDIR_DIFF is the wind direction _WINDIR of the current second minus the wind direction _WINDIR of the last second, and the calculated difference is unified to the interval [-180, 180).

[0082] The wind speed standard deviation std_WINSPD in the past four seconds is the standard deviation of the wind speed _WIN_DIR data in the past four seconds.

[0083] The calculation process of the wind direction fluctuation WINHEAD_fluct in the past four seconds is: mapping the wind direction vector data in the past four seconds to the unit circle, calculating the distance d from the center of the four points to the center of the unit circle, and taking 1-d as the wind direction fluctuation value in the past four seconds.

[0084] It should be noted that the past four seconds refer to the previous four seconds including the current second, for example, the current second is the 6th second, and the past four seconds refer to the 3rd, 4th, 5th, and 6th seconds.

[0085] Through the calculation of the pilot operation amplitude mean value of each second of all flight aircraft and the wind field related variables, the characteristics of the wind field data are further mined.

[0086] Further, in step 3

[0087] For each second of data before the aircraft lands, based on the wind field related variables, the K-Nearest Neighbors (KNN) method is used to search for matching data in the flight data of all flights, forming a data set, and the specific operation includes:

[0088] For each second of data before the aircraft lands, the Euclidean distance between all flight data of the aircraft is calculated using wind speed, wind direction and heading angle, wind speed change value, wind direction change value, wind speed standard deviation in the past four seconds, and wind direction fluctuation data in the past four seconds. K data (such as 100 or 500 data) with the closest Euclidean distance are selected, which are considered to have similar wind field characteristics to form a data set.

[0089] Among them, each second corresponds to a data set, and based on each data set, the quantified index of the pilot operation amplitude fluctuation degree corresponding to each second can be obtained.

[0090] Among them, the K-Nearest Neighbors (KNN) method used in the embodiment of the application is a prior art method, which will not be described here.

[0091] Further, the step 3 of calculating the quantified index of the pilot operation amplitude fluctuation degree based on the data set comprises:

[0092] The 10% quantile and the 90% quantile of the pilot's pitch operation amplitude mean in the data set are calculated, and the interquartile range IQR is obtained by the difference between the two, as the quantified index of the pilot's pitch operation amplitude fluctuation degree corresponding to a second;

[0093] The 10% quantile and the 90% quantile of the pilot's roll operation amplitude mean in the data set are calculated, and the interquartile range IQR is obtained by the difference between the two, as the quantified index of the pilot's roll operation amplitude fluctuation degree corresponding to a second.

[0094] The quantified index of the pilot's pitch operation amplitude fluctuation degree corresponding to each second or the quantified index of the pilot's roll operation amplitude fluctuation degree corresponding to each second is obtained by sequentially calculating each data set.

[0095] Among them, the pilot's pitch operation amplitude mean or roll operation amplitude mean in a data set is sequentially sorted from small to large, and the position corresponding to the 10% data value is the 10% quantile of the pilot's pitch operation amplitude mean or roll operation amplitude mean, and the position corresponding to the 90% data value is the 90% quantile of the pilot's pitch operation amplitude mean or roll operation amplitude mean.

[0096] The quantified index of the pitch operation amplitude fluctuation degree and the quantified index of the roll operation amplitude fluctuation degree provide dependent variables for the construction of the turbulence index.

[0097] Further, in step 4,

[0098] The quantified index of the pilot operation amplitude fluctuation degree is fitted based on the wind field related variables, and a semi-parametric regression model is constructed, which comprises:

[0099] The quantification index of the fluctuation degree of the pilot's pitch operation amplitude is fitted by taking height as the nonlinear part and taking wind speed, the angle between wind direction and heading, wind speed change value, wind direction change value, wind speed standard deviation in the past four seconds, and wind direction fluctuation in the past four seconds as the linear part, and a semi-parametric regression model of the pitch operation is constructed.

[0100] The quantification index of the fluctuation degree of the pilot's roll operation amplitude is fitted by taking height as the nonlinear part and taking wind speed, the angle between wind direction and heading, wind speed change value, wind direction change value, wind speed standard deviation in the past four seconds, and wind direction fluctuation in the past four seconds as the linear part, and a semi-parametric regression model of the roll operation is constructed.

[0101] The general expression of the semi-parametric regression model can be expressed as:

[0102] IQR = X T β + f(H)

[0103] wherein X is a vector of wind field related variables, which is different according to whether the operation considered is pitch or roll, X T β is a linear combination of variables, β is a regression coefficient variable, H is the modified radio altitude, and f(H) is a function of H.

[0104] The model reveals the numerical relationship between the IQR data and the wind field characteristic related variables and the height.

[0105] Further, in the step 5,

[0106] Based on the semi-parametric regression model, a turbulence index calculation formula is obtained, and the specific operation includes:

[0107] The influence of height on the semi-parametric regression model of the pitch or roll operation is eliminated (i.e., f(H) in the expression of the semi-parametric regression model is deleted), and a turbulence index calculation formula for the pilot's pitch or roll operation is obtained.

[0108] For the fitted semi-parametric regression model of the pitch or roll operation, the coefficients of the linear part can be obtained, and the result X T β of the linear part is directly solved, that is, the obtained turbulence index calculation formula for the pilot's pitch or roll operation is obtained.

[0109] According to the turbulence index, the influence of the wind field on the pilot's operation in the approach phase is investigated, and the influence of turbulence on flight is determined.

[0110] The turbulence index calculation formula is used to calculate the QAR data of the new flight, specifically: the wind field related variables are calculated based on the QAR data of the new flight, then the wind field related variables are input into the turbulence index calculation formula, and the turbulence index is obtained, so as to test the influence of the wind field on the pilot operation in the approach phase of the new flight, and check the related effect of the model.

[0111] Taking two flights with different air flows as examples, the turbulence index is calculated by the method. Figure 2 For the turbulence index of the flight with relatively stable air flow, the turbulence index value range and the flight warning value are small, and for the turbulence index of the flight with relatively turbulent air flow, the turbulence index value range and the flight warning value are large. Figure 3 For the turbulence index of the flight with relatively turbulent air flow, the turbulence index value range and the flight warning value are large. When identifying the degree of air flow confusion, a certain accuracy is achieved, and the warning effect in actual application is enhanced.

[0112] The embodiment of the application also provides a determination device for the turbulence index of the landing of the aircraft, which is used to execute the steps in the determination method for the turbulence index of the landing of the aircraft, and the system comprises:

[0113] A data acquisition unit is configured to acquire QAR data of all flight aircrafts of the same type in the low-altitude and near-ground phase of the same airport;

[0114] A first calculation unit is configured to calculate the pilot operation amplitude mean value and the wind field related variable of each flight aircraft per second based on the QAR data;

[0115] A second calculation unit is configured to search for matched data in the flight data of all flight aircrafts based on the wind field related variable by using the K-neighbor method for each second of data before the aircraft lands, form a data set, and calculate a quantitative index of the pilot operation amplitude fluctuation degree based on the data set;

[0116] A model construction unit is configured to fit the quantitative index of the pilot operation amplitude fluctuation degree based on the wind field related variable, and construct a semi-parametric regression model;

[0117] A turbulence index determination unit is configured to obtain a turbulence index calculation formula based on the semi-parametric regression model.

[0118] The embodiment of the application provides a computer readable storage medium, which stores at least one computer executable program, and the at least one program is executed by the computer to make the computer execute the steps in the determination method for the turbulence index of the landing of the aircraft,

[0119] As shown in the figure, the steps specifically include: Figure 1

[0120] ​Step 1, collect QAR data of all flights of the same aircraft model at low altitude and near ground stage in the same airport;

[0121] Step 2, calculate the mean value of pilot operation amplitude and wind field related variables of each flight per second based on QAR data;

[0122] Step 3, for each second of data before the aircraft lands, respectively based on the wind field related variables, search the matching data in the flight data of all flights by using K-Nearest Neighbors (KNN) method, form a data set, and calculate the quantification index of the pilot operation amplitude fluctuation degree based on the data set;

[0123] Step 4, fitting the quantification index of the pilot operation amplitude fluctuation degree based on the wind field related variables, and constructing a semi-parametric regression model;

[0124] Step 5, based on the semi-parametric regression model, obtain the turbulence index calculation formula.

[0125] The embodiment of the application provides a computer device, comprising a memory and a processor, the memory stores a computer program, characterized in that the processor executes the computer program to realize the steps in the determination method of the aircraft landing turbulence index, as shown in Figure 1 The specific steps include the following steps:

[0126] Step 1, collect QAR data of all flights of the same aircraft model at low altitude and near ground stage in the same airport;

[0127] Step 2, calculate the mean value of pilot operation amplitude and wind field related variables of each flight per second based on QAR data;

[0128] Step 3, for each second of data before the aircraft lands, respectively based on the wind field related variables, search the matching data in the flight data of all flights by using K-Nearest Neighbors (KNN) method, form a data set, and calculate the quantification index of the pilot operation amplitude fluctuation degree based on the data set;

[0129] Step 4, fitting the quantification index of the pilot operation amplitude fluctuation degree based on the wind field related variables, and constructing a semi-parametric regression model;

[0130] Step 5, based on the semi-parametric regression model, obtain the turbulence index calculation formula.

[0131] Those skilled in the art can understand that all or part of the processes in the above-mentioned embodiment methods can be completed by instructing the relevant hardware through a computer program. The computer program can be stored in a non-volatile computer readable storage medium, and when the computer program is executed, the processes of the above-mentioned embodiments of the methods can be included. Any reference to memory, storage, database or other medium used in the embodiments provided in the present application can include non-volatile and / or volatile memory. Non-volatile memory can include read-only memory (ROM), programmable ROM (PROM), electrically programmable ROM (EPROM), electrically erasable programmable ROM (EEPROM) or flash memory. Volatile memory can include random access memory (RAM) or external cache memory. As an illustration but not limitation, RAM is available in various forms, such as static RAM (SRAM), dynamic RAM (DRAM), synchronous DRAM (SDRAM), double data rate SDRAM (DDR SDRAM), enhanced SDRAM (ESDRAM), synchronous link (Synchlink) DRAM (SLDRAM), memory bus (Rambus) direct RAM (RDRAM), direct memory bus dynamic RAM (DRDRAM), and memory bus dynamic RAM (RDRAM), etc.

[0132] The above technical solutions are only one embodiment of the present application. For those skilled in the art, on the basis of the principles disclosed in the present application, various types of improvements or modifications can be easily made, and are not limited to the technical solutions described in the above embodiments of the present application. Therefore, the above description is only preferred, and is not limited in nature.

Claims

1. A method of determining an aircraft landing turbulence index, characterized in that, Specifically comprising the following steps: Step 1, collect QAR data of all flights of the same aircraft type at the same airport in the low-altitude and near-ground stage; Step 2, based on the QAR data, calculate the pilot operation amplitude mean value and the wind field related variable of each flight per second respectively; Step 3, for each second of data before the aircraft lands, based on the wind field related variable, use the K nearest neighbor method to search for matching data in the flight data of all flights, form a data set, and calculate the quantification index of the pilot operation amplitude fluctuation degree based on the data set, the specific operation including: Calculate the 10% quantile and 90% quantile of the pilot's pitch operation amplitude mean value in the data set, and obtain the interquartile range IQR by the difference between the two, as the quantification index of the pilot's pitch operation amplitude fluctuation degree corresponding to a second; Calculate the 10% quantile and 90% quantile of the pilot's roll operation amplitude mean value in the data set, and obtain the interquartile range IQR by the difference between the two, as the quantification index of the pilot's roll operation amplitude fluctuation degree corresponding to a second; Calculate the quantification index of the pilot's pitch operation amplitude fluctuation degree corresponding to a second or the quantification index of the pilot's roll operation amplitude fluctuation degree corresponding to a second based on each data set in turn; Step 4, fitting the quantification index of the pilot operation amplitude fluctuation degree based on the wind field related variable, constructing a semi-parametric regression model, the specific operation including: Fitting the quantification index of the pilot's pitch operation amplitude fluctuation degree with height as the nonlinear part and wind speed, wind direction and heading angle, wind speed change value, wind direction change value, wind speed standard deviation in the past four seconds and wind direction fluctuation in the past four seconds as the linear part, constructing a semi-parametric regression model for pitch operation; Fitting the quantification index of the pilot's roll operation amplitude fluctuation degree with height as the nonlinear part and wind speed, wind direction and heading angle, wind speed change value, wind direction change value, wind speed standard deviation in the past four seconds and wind direction fluctuation in the past four seconds as the linear part, constructing a semi-parametric regression model for roll operation; Step 5, based on the semi-parametric regression model, obtaining the turbulence index calculation formula.

2. The method of claim 1, wherein, In step 1 of the semi-parametric regression model, the QAR data includes: height, flight stage, heading, wind direction, wind speed, automatic pilot state, captain's pitch operation amplitude, co-pilot's pitch operation amplitude, captain's roll operation amplitude and co-pilot's roll operation amplitude.

3. The method of claim 2, wherein, In step 2, Pilot's pitch operation amplitude mean value = sum of captain's or co-pilot's pitch operation amplitude collected per second / total number of captain's or co-pilot's pitch operation amplitude per second; Pilot's roll operation amplitude mean value = sum of captain's or co-pilot's roll operation amplitude collected per second / total number of captain's or co-pilot's roll operation amplitude per second.

4. The method of claim 3, wherein, In step 2, the wind field related variable includes wind direction and heading angle, wind speed change value, wind direction change value, wind speed standard deviation in the past four seconds and wind direction fluctuation in the past four seconds.

5. The method of claim 4, wherein, The operation in step 5 includes: Eliminate the influence of height from the semi-parametric regression model of pitch operation to obtain the turbulence index calculation formula for the pilot's pitch operation; The influence of height is removed from the semi-parametric regression model of roll operation to obtain a turbulence index calculation formula for the roll operation of the pilot.

6. An apparatus for determining an aircraft landing turbulence index, characterized by The system comprises: a data acquisition unit configured to acquire QAR data of all flights of the same aircraft type at a same airport in a low-altitude and near-ground phase; a first calculation unit configured to calculate a mean value of a pilot operation amplitude and a wind field related variable of each flight per second based on the QAR data; a second calculation unit configured to search for matching data in flight data of all flights based on the wind field related variable and using a K-neighbor method for each second of data before the aircraft lands, form a data set, and calculate a quantification index of the pilot operation amplitude fluctuation degree based on the data set; a model construction unit configured to fit the quantification index of the pilot operation amplitude fluctuation degree based on the wind field related variable and construct a semi-parametric regression model; a turbulence index determination unit configured to obtain a turbulence index calculation formula based on the semi-parametric regression model.

7. A computer-readable storage medium, characterized in that, The computer readable storage medium stores at least one program executable by a computer, and the at least one program causes the computer to execute the steps in the method for determining a turbulence index of an aircraft landing according to any one of claims 1-5 when executed by the computer.

8. A computer device, comprising: The system comprises a memory and a processor, and the memory stores a computer program, and the processor implements the steps in the method for determining a turbulence index of an aircraft landing according to any one of claims 1-5 when executing the computer program.

Images