CN114282792A - Flight landing quality monitoring and evaluating method and system - Google Patents

Flight landing quality monitoring and evaluating method and system Download PDF

Info

Publication number
CN114282792A
CN114282792A CN202111559889.3A CN202111559889A CN114282792A CN 114282792 A CN114282792 A CN 114282792A CN 202111559889 A CN202111559889 A CN 202111559889A CN 114282792 A CN114282792 A CN 114282792A
Authority
CN
China
Prior art keywords
landing
data
measurement
flight
heavy
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Pending
Application number
CN202111559889.3A
Other languages
Chinese (zh)
Inventor
俞力玲
赵新斌
尚家兴
郑林江
周秀婷
陈红年
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
Chongqing University
China Academy of Civil Aviation Science and Technology
Original Assignee
Chongqing University
China Academy of Civil Aviation Science and Technology
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by Chongqing University, China Academy of Civil Aviation Science and Technology filed Critical Chongqing University
Priority to CN202111559889.3A priority Critical patent/CN114282792A/en
Publication of CN114282792A publication Critical patent/CN114282792A/en
Pending legal-status Critical Current

Links

Images

Landscapes

  • Traffic Control Systems (AREA)

Abstract

The invention provides a flight landing quality monitoring and evaluating method and system, and belongs to the field of data mining, aviation safety and risk evaluation. The method utilizes original QAR data of the airplane to obtain a Measurement index related to the heavy landing, and utilizes the Measurement index to obtain a heavy landing risk value. The method comprises the following steps: (1) collecting original QAR data of an airplane; (2) extracting flight parameter Measurement values, namely Measurement data, from original QAR data; (3) preprocessing the data; (4) and constructing a heavy landing risk evaluation model to obtain a heavy landing risk value. By using the method and the device, an airline company can know the flight heavy landing risk degree and explore the risk reason of the flight heavy landing risk degree, so that the massive flight data management and the automatic evaluation are facilitated, and valuable information such as risk change trend and the like can be displayed visually.

Description

Flight landing quality monitoring and evaluating method and system
Technical Field
The invention belongs to the field of data mining, aviation safety and risk assessment, and particularly relates to a flight landing quality monitoring and assessment method and system.
Background
In the modern society developed in the aviation industry, aviation safety is always one of the most important topics in the aviation field. According to statistics of the Boeing company on the data of the major flight safety accidents from 2009 to 2019, the final landing stage is a stage where the flight safety accidents are very easy to occur. The main reason is that the aircraft requires a lot of pilot operations and complex environmental changes at this stage, and misjudgment or misoperations by the pilot may cause serious consequences such as causing safety incidents. Therefore, research and analysis on the land stage have important significance in the aviation safety field.
Among the landing safety events, a heavy landing is a typical landing safety event, which is an event in which the impact load (generally expressed by vertical acceleration) generated by the landing gear and the ground exceeds a specified limit at the moment of landing the aircraft. In the safety events in the landing stage, heavy landing is one type of unsafe events which occur frequently, and the total number of unsafe events in the landing stage is about 20% since the heavy landing unsafe event 125 occurs in China civil aviation in 2006-2011. In 1993-2002, 385 accidents occurred globally, wherein 2 airplanes were damaged due to heavy landing, 47 airplanes were seriously damaged, and 11 airplanes were slightly damaged. As a typical unsafe landing risk event, heavy landing not only brings bad flight experience to passengers and damages the image of an airline company, but also accelerates the fatigue damage and even breakage of the whole structure of wings, landing gears, engines and airplanes, increases the occurrence probability of landing safety accidents, brings huge economic loss to the airline company, and causes disastrous accident consequences when the situation is serious, thereby threatening the life safety of passengers.
The current research on heavy landing is mainly divided into two categories: the first category is heavy landing studies based on management and psychology, and the second category is heavy landing studies based on QAR (quick access recorder, airborne flight data recording device with protection) data.
The first group of researchers are generally professionals in the field of aviation, who have a deeper understanding of the field of civil aviation, mainly from the aspects of management and psychology, to analyze the cause of heavy landing and prevention measures. The problems of airplane re-landing are analyzed and considered by stone 28156: weather reasons, pilot operating technical reasons, and other reasons. Lijiahua summarizes the causes of heavy landing as: unstable approach, improper flight control and misjudgment caused by environment or special weather conditions. For the reasons of heavy landing, the authors respectively give improvement suggestions from the aspects of pilot control technical points, attention allocation, pilot morphology, weather condition forecast of landing airports, and the like.
The second category is QAR-based heavy landing research, which is called data-driven research because it mainly uses machine learning and risk assessment models to analyze and mine QAR data to find the occurrence cause or probability of heavy landing. Compared with the research based on management and psychology, the research speaks in data, so the result is more objective, and the flight expert can be helped to discover some new useful information, and the research of heavy landing based on QAR data is gradually becoming the mainstream at present. A three-layer BP neural network is designed for predicting heavy landing in Caochitong and the like, and scholars research a heavy landing prediction model based on SVM. The Tongtao et al use a deep learning framework to solve the problem of heavy landing, and provide a heavy landing prediction framework based on a long-short term memory network aiming at the time sequence characteristics of QAR data. These methods are all predictions made before a heavy landing occurs, so that it is desirable to avoid the occurrence of a heavy landing as much as possible. In the aspect of risk assessment, the Wan proposes a QAR-based quantitative assessment model of the risk of heavy landing, which defines the risk of heavy landing as the product of the probability of occurrence of heavy landing and the severity of the occurrence of heavy landing event, and further perfects the risk model on the basis of the model, and simultaneously considers the three events of heavy landing, drift out of the runway and tail wiping.
However, the existing risk assessment method only considers the risk assessment of the heavy landing in terms of both the occurrence probability and the occurrence severity, and does not explore some characteristic parameters of a specific time point or a specific time period causing the risk.
Disclosure of Invention
The invention aims to solve the problems in the prior art, and provides a flight landing quality monitoring and evaluating method and system, which are convenient for an airline company to know the flight heavy landing risk degree and explore the risk reason, convenient for realizing massive flight data management and automatic evaluation, and capable of displaying valuable information such as risk change trend through visualization.
The invention is realized by the following technical scheme:
in a first aspect of the invention, a flight landing quality monitoring and evaluating method is provided, which obtains a Measurement index related to heavy landing by using original QAR data of an airplane, and obtains a heavy landing risk value by using the Measurement index.
The invention is further improved in that:
the method comprises the following steps:
(1) collecting original QAR data of an airplane;
(2) extracting flight parameter Measurement values, namely Measurement data, from original QAR data;
(3) preprocessing the data;
(4) and constructing a heavy landing risk evaluation model to obtain a heavy landing risk value.
The invention is further improved in that:
the operation of the step (2) comprises the following steps:
flight parameter measurements, namely, Measurement data, are extracted from the original QAR data using Measurement software.
The invention is further improved in that:
the operation of the step (3) comprises:
(31) preprocessing original QAR data;
(32) preprocessing the Measurement data.
The invention is further improved in that:
the operation of the step (4) comprises the following steps:
(41) extracting data related to heavy landing from the preprocessed Measurement data to serve as Measurement indexes, and drawing a box line graph of each Measurement index;
(42) establishing a reference range of each Measurement index, and determining the weight of each Measurement index;
(43) and obtaining a heavy landing risk value.
The invention is further improved in that:
the operation of drawing the box line graph of each Measurement index in the step (41) includes:
selecting a time period of 30 seconds forward and 20 seconds backward from the grounding time, and taking a total of 50 seconds as a landing interval;
for each flight segment, firstly extracting a change curve of a vertical load VRTG in a landing interval, then acquiring a peak value of the change curve, if the peak value is greater than a threshold value, judging that the flight segment is a heavy landing flight segment, and if the peak value is less than or equal to the threshold value, judging that the flight segment is a non-heavy landing flight segment;
and drawing a box line graph of each Measurement index, wherein the box line graph comprises a non-heavy landing leg and a heavy landing leg.
The invention is further improved in that:
the operation of step (42) comprises:
selecting a Measurement index which is obvious in distinguishing between heavy landing flight legs and non-heavy landing flight legs according to the box diagram;
distinguishing a remarkable Measurement index for each heavy landing flight leg and a non-heavy landing flight leg, obtaining a range of 90% of Measurement values of the Measurement index in a landing interval, and taking the range as a reference range of the Measurement index; and setting the weight of the Measurement index according to the capability of distinguishing the heavy landing leg from the non-heavy landing leg, wherein the value range of the weight is [ 0-1 ].
The invention is further improved in that:
the operation of said step (43) comprises:
and calculating to obtain a heavy landing Risk value Risk of one flight segment by using the following formula:
Figure BDA0003420195790000041
where M represents the set of all Measurement indicators used to calculate the heavy landing risk value, MiIs the ith Measurement index, I {. is a symbolic function, which represents: when the condition in brackets is established, I { } is 1, otherwise, I { } is 0, and wiIs the weight of the ith Measurement index;
the map rule indicates that the calculation rule is satisfied, which is as follows: the calculation rule includes three value types: left, Right and Both, where Left indicates that the measurement values are Left-shifted out of the reference range, Right indicates that the measurement values are Right-shifted out of the reference range, and Both indicates that the measurement values are Left-shifted out of the reference range or Right-shifted out of the reference range.
In a second aspect of the present invention, a flight landing quality monitoring and evaluating system is provided, the system comprising:
the device comprises an acquisition unit, a data acquisition unit and a data processing unit, wherein the acquisition unit is used for acquiring original QAR data of the airplane;
the extracting unit is connected with the collecting unit and used for extracting flight parameter measured values, namely Measurement data, from the original QAR data;
the preprocessing unit is respectively connected with the acquisition unit and the extraction unit and is used for preprocessing data;
the risk value calculation unit is connected with the preprocessing unit and used for constructing a heavy landing risk evaluation model to obtain a heavy landing risk value;
a visualization unit: and the risk value calculation unit is connected with the acquisition unit, the extraction unit, the preprocessing unit and the risk value calculation unit respectively and is used for visually displaying the data.
Compared with the prior art, the invention has the beneficial effects that: by using the method and the device, an airline company can know the flight heavy landing risk degree and explore the risk reason of the flight heavy landing risk degree, so that the massive flight data management and the automatic evaluation are facilitated, and valuable information such as risk change trend and the like can be displayed visually.
Drawings
FIG. 1 is a diagram of the effect of different interpolation methods;
FIG. 2 is a wind speed and direction diagram;
FIG. 3 example Measurement table contents;
4-1Touchdown Gate 1(200 feet height) to 50 feet phase maximum descent rate box plot of the aircraft;
4-2Touchdown Gate 1(200 feet height) to 50 feet phase minimum descent rate box plot of the aircraft;
FIG. 5-1 curve IVV for a heavy landing; (ii) a
FIG. 5-2 curve IVV for a non-heavy landing;
FIG. 6-150 ft to ground mean rate of descent boxplot comparison results;
FIG. 6-250 ft to ground maximum rate of descent boxplot comparison results;
FIGS. 6-350 ft to ground minimum rate of descent boxplot comparison results;
graph 750 ft to ground time boxed plot comparison results;
FIG. 8 minimum VRTG values for the landing phase;
FIG. 9 shows the comparison of the maximum pitch angle boxplots at the landing stage;
FIG. 10 comparison of maximum pitch angle boxplots at the moment of grounding;
comparison of the plot 1150 ft to the maximum pitch angle box plot at the ground phase;
FIG. 12 is a comparison of the maximum absolute roll box plots at the time of grounding;
FIG. 1350 ft to ground phase maximum absolute roll angle box plot;
FIG. 14-1 is a diagram of an energy bin at the moment of grounding;
14-250 ft time energy boxplot;
FIG. 15-1 is a plot of the ground speed of the aircraft at the moment of grounding;
FIG. 15-250 ft time aircraft ground speed contour plot;
FIG. 16-1 is a plot of the maximum groundspeed box during the landing phase;
FIG. 16-2 plot of maximum airspeed box during the landing phase;
FIG. 17 comparison of maximum lateral acceleration boxplots during the landing phase;
graph 1850 ft to ground phase minimum side wind box plot comparison results;
FIG. 19 is a plot of risk values for heavy landing versus boxed plot;
FIG. 20 is a scatter diagram of the relationship between the heavy landing risk and the VRTG peak value;
FIG. 21 is an overall block diagram of the system of the present invention;
FIG. 22 is a block diagram of an implementation of the system of the present invention;
FIG. 23 is a Teblueau based data drill analysis;
FIG. 24-1 case analysis of risk of heavy landing;
FIG. 24-2 case analysis of risk of heavy landing;
FIG. 24-3 case analysis of risk of re-landing;
FIG. 24-4 case analysis of risk of re-landing;
FIG. 25 is a block diagram of the steps of the method of the present invention.
Detailed Description
The invention is described in further detail below with reference to the accompanying drawings:
the characteristic parameters have great practical significance for researching the heavy landing risk degree and the risk reason. According to the method, the characteristic parameters of certain time points or time periods of the Measurement data are utilized, and the degree of deviation of the characteristic from a normal range is considered to construct the heavy landing risk assessment model, so that an airline can conveniently know the heavy landing risk degree of the flight and explore the risk reason of the heavy landing risk degree. The system is toolized on the basis, so that massive flight data management and automatic evaluation can be realized, and valuable information such as risk change trend and the like can be displayed visually.
The invention provides a flight landing quality monitoring and evaluating method, as shown in fig. 25, the method comprises the following steps:
(1) collecting original QAR data of an aircraft:
the original QAR data of the airplane completely records various parameters of the whole flight stage of the airplane, including state parameters of the airplane, control command parameters of a pilot, external environment parameters and the like. Table 1 lists some typical QAR data.
Figure BDA0003420195790000061
Figure BDA0003420195790000071
TABLE 1
(2) Flight parameter measurements, namely Measurement data, are extracted from the raw QAR data:
the flight parameter measurements can be extracted from the original QAR data using the existing tool, "Measurement", which is referred to as Measurement data.
The Measurement is a data Measurement and acquisition tool developed by Teledyne and oriented to the field of aviation application, and can extract flight parameter Measurement values interested by users from original QAR data through operations such as decoding, calculation, Measurement, aggregation and the like. Flight parameter measurements extracted by Measurement are classified into the following 4 classes:
information Measurements: such measures information includes basic information of a certain flight, such as model, takeoff airport ICAO code, landing airport ICAO code, flight number, and the like.
Scalar Measurements: such Measurements are Measurements taken from a single sampling instant (unit: second) of a leg, such as: touchdown time.
Aggregate measures: the measurement information is a measurement value obtained by aggregating according to a predefined aggregation method from a certain predefined time range, wherein the time range comprises two time points, flight phases or whole flight phases, and the aggregation method comprises sum, max, min, average and the like.
Combined Aggregate measures: such Measurements are obtained by aggregating measurement objects that are identical in a plurality of ranges, such as all cruise phases of a flight.
The Measurement has predefined many common Measurement values, and it supports the user to extend on the existing Measurement definition to get the self-defined Measurement.
(3) Preprocessing data:
firstly, preprocessing original QAR data, for example, in order to obtain the sizes of axial wind and lateral wind of an airplane, three parameters of wind speed, wind direction and magnetic heading in the original QAR data need to be processed to obtain two new flight parameters of the axial wind and the lateral wind, and the processing of the original QAR data with different frequencies and the parameter digitization of state types are also needed;
after the pre-processing of the original QAR data is completed, the Measurement data is pre-processed.
The operation of the step (3) is specifically as follows:
(31) preprocessing the original QAR data:
most QAR data is of a numerical type, such as radio altitude, airspeed, ground speed, pitch angle, roll-off rate, vertical load, etc., while other parameters are discrete state variables. For example, the landing gear state mainly includes two states of AIR and GROUND, which correspond to the opening and closing of an AIR-to-GROUND electric door of the landing gear, AIR represents that the landing gear is in the AIR (not grounded), and GROUND represents that the landing gear is grounded. In order to visualize the landing gear state of an aircraft, two states of AIR and GROUND need to be converted into numerical types, and in order to make the result as intuitive as possible, the invention converts the AIR state into 1 and the GROUND state into 0. Similar operations may be performed for other discrete status parameters (AP _ EGD1, AP _ EGD2, ATHR _ EGD, FLAP _ LEVEL1, FLAP _ LEVEL 2).
In addition, the sampling frequencies of different QAR data are different, and in order to visualize the QAR data change curve in a fine-grained manner and to perform comparative analysis on different parameter curves in the same graph, the invention needs to perform interpolation processing on some data with lower sampling frequency and expand the sampling frequency to the highest sampling frequency (8 Hz). Different interpolation means are required for different types of data. For example, for a numerical parameter such as the radio altitude, a linear interpolation method may be adopted, and since the sampling frequency of the radio altitude itself is 4Hz, to expand to 8Hz, it is only necessary to interpolate the average value (x1+ x2)/2 between two adjacent data points x1, x 2. However, for the landing gear state, which is a discrete state parameter, it is not suitable to adopt a linear interpolation method, assuming that the AIR state corresponds to 1 and the GROUND state corresponds to 0, an intermediate value of 0.5 may be obtained after interpolation, and the intermediate value does not correspond to the landing gear state, for such parameters, the present invention adopts a "covering" interpolation method, i.e. a value of a previous point nearest to the point is covered by a previous point (i.e. the current point and the previous point nearest to the point have the same value), and the effects of the two kinds of interpolation are shown in fig. 1. In summary, for QAR parameters of different types and different sampling frequencies, a corresponding interpolation method needs to be adopted for processing according to specific situations.
In addition to the above data transformation and interpolation process, some parameters, including two parameters of wind speed and wind direction, need to be highly transformed. Research shows that wind shear in the landing stage is also one of the causes of unsafe landing events, however, the wind shear is relative to the flight direction of the airplane (upwind to downwind, left wind to right wind, etc.), and it is difficult to obtain exact information of the wind shear directly from the wind speed and direction parameters, so that the effect of wind on the airplane needs to be calculated by combining the magnetic heading and the wind speed and direction of the airplane. The calculation method comprises the following steps: firstly, calculating an included angle theta between a wind direction and an airplane magnetic heading, wherein beta is the wind direction and is the alpha magnetic heading, and then calculating components of the wind speed in parallel and perpendicular to the airplane heading by utilizing a trigonometric function relationship, wherein the included angle theta is beta-alpha, and the component is as follows:
Figure BDA0003420195790000091
wherein WINSPDRepresenting wind speed, WINCRSIndicating crosswind, positive values indicating left crosswind, negative values indicating right crosswind, WINALGThe axial wind is shown, positive values indicate downwind, negative values indicate upwind, and the calculation diagram is shown in fig. 2.
(32) Preprocessing the Measurement data:
the method carries out data preprocessing operation based on the Measurement data, including data cleaning, missing value filling, data association and the like. For preprocessing the field of the AircraftType, firstly, analyzing the field of the Flight table (the Flight table is derived from a Measurement database and is a field containing various Flight parameter fields (such as Flight time, airplane model, namely the AircraftType) to find that the field can be provided with two names corresponding to the same airplane type, such as B737 and 737, and for the field, adopting a processing mode of unifying the field value into a standard value, such as modifying all '737' into 'B737', and for the missing processing of the field content, in the Measurement data, the situation that a part of the Flight segment of the fields related to take-off and landing airports is empty exists, and when performing the airport related analysis on the data, unifying the data into other airport categories.
The Measurement table stores the Measurement value specific information of multiple legs, and the content of the Measurement value specific information is shown in fig. 3, and includes fields such as a flight id, a Measurement code, and a Measurement desc, and table 2 lists the field names and specific meanings in the Measurement table. The Flight ID field specifies which Flight segment the Measurement value comes from, so that the Flight table and the Measurement table can be associated through the Flight ID, and all Measurement values of a certain Flight segment can be obtained.
Figure BDA0003420195790000092
Figure BDA0003420195790000101
TABLE 2
In order to determine whether the MeasurementCode can be used as the unique identifier of a certain measurement mode, the invention analyzes the corresponding relationship between the MeasurementCode and measurementdescr, MeasurementName, measuremenrange, RangeType, and agregate, and the result is shown in table 3. It can be seen that there is a one-to-one correspondence between MeasurementCode and measurementdescr, but not with other fields. However, for the same MeasurementCode, the measurementdescr, the MeasurementName, the measuremenrange, the RangeType, and the Aggregate are all unique, so that the MeasurementCode can be used as a unique identifier of a certain measurement mode.
Figure BDA0003420195790000102
TABLE 3
According to the method, part of fields related to the land security are selected, the MeasurementValue measurement value is extracted, the box line graph is used for displaying, and obvious abnormality exists in part of the measurement value. Taking the descent rate (IVV) during the landing phase as an example, as shown in fig. 4-1 and 4-2, fig. 4-1 is a box plot of the maximum descent rate of the aircraft during the Touchdown Gate 1(200 feet in height) to 50 feet phase, and fig. 4-2 is a box plot of the minimum descent rate, and it can be seen that the maximum and minimum descent rates include values (greater than 10000 or less than-10000) that deviate significantly from the normal range of values.
In addition to this parameter, it was found that similar outlier conditions exist for some of the other parameters. In order to solve the problems, the invention provides a rule-based abnormal value filtering method, and gives a normal value range of a main Measurement index by combining with expert experience. When the Measurement data is cleaned, the Measurement values exceeding the range are removed, and the data with higher quality is obtained.
(4) Constructing a heavy landing risk assessment model:
(41) extracting data related to heavy landing from the preprocessed Measurement data to serve as Measurement indexes, and drawing a box line graph of each Measurement index, wherein the box line graph comprises two parts of non-heavy landing and heavy landing:
firstly, in order to analyze a heavy landing event, the method extracts a landing interval in preprocessed QAR data, and selects 30 seconds from the grounding moment to the front and 20 seconds from the back as the landing interval, wherein the total time is 50 seconds. The method comprises the steps of firstly extracting a change curve of a vertical load (VRTG) of a landing interval for each flight segment, then obtaining a peak value of the VRTG curve, judging that the flight segment is subjected to heavy landing if the peak value is larger than a threshold value, namely a heavy landing flight segment, judging that the flight segment is not subjected to heavy landing if the peak value is smaller than or equal to the threshold value, namely a non-heavy landing flight segment, dividing the flight segment into the heavy landing flight segment and the non-heavy landing flight segment through the threshold value, and dividing the box line graph into the non-heavy landing part and the heavy landing part according to the threshold value when the box line graph is drawn below.
In this embodiment, analysis is performed based on QAR data, and all models are a320, and include 53 heavy landing legs (VRTG peak value is greater than 1.8G) and 101 non-heavy landing legs.
Since the occurrence of a heavy landing is generally closely related to the descent rate of the aircraft, the present invention further performs a visual analysis of the descent rate (IVV) curve, as shown in FIGS. 5-1 and 5-2. It can be seen that IVV for the heavy landing leg fluctuates more than IVV for the non-heavy landing leg before the 50 foot altitude is reached. In addition, the IVV curve characteristics of the two show a significant difference between the 50ft to ground contact interval, as shown in FIGS. 5-1 and 5-2, where the heavy landing leg IVV curve is "convex downward" rather than "convex upward". This feature indicates that the re-landing leg does not control the descent rate of the aircraft in time after the aircraft enters the 50ft altitude because the aircraft is quickly grounded, resulting in a large vertical load. And after the non-heavy landing enters 50ft height, the descent rate control is obvious, so that the descent rate of the airplane grounded is low, and the occurrence of heavy landing is avoided. Timely and effective control of the descent rate of an aircraft as it enters a 50ft altitude is critical to avoiding a heavy landing.
Extracting data related to heavy landing from the preprocessed Measurement data to serve as a Measurement index:
the Measurement indexes obtained through system carding comprise two types: one type is the Scalar Measurement and the other type is the Aggregate Measurement. Table 4 shows the Scalar Measurement indicators related to heavy landing, and table 5 shows the Aggregate Measurement indicators related to heavy landing (these indicators are directly extracted from QAR data, and named as different Measurement indicators, such as Airspeed Touchdown, meaning the Airspeed value of the aircraft at the landing time, according to different selected parameters and different time points).
Figure BDA0003420195790000111
Figure BDA0003420195790000121
TABLE 4
Figure BDA0003420195790000122
Figure BDA0003420195790000131
TABLE 5
Based on the sorted Measurement index results, 49 Measurement indexes are finally selected for extraction, and box line graphs are adopted to compare differences of different indexes of the heavy landing segment and the non-heavy landing segment, so that the method assists in finding out which indexes have important reference significance for calculating the heavy landing risk.
Fig. 6-1 through 6-3 illustrate boxplots of the avg _ ivv _50ft _ td, max _ ivv _50ft _ td, and min _ ivv _50ft _ td (average, maximum, minimum descent rate from 50ft to ground) indicators, and it can be seen that the descent rate for the Hard landing (Hard landing) leg is significantly higher than the descent rate for the non-Hard landing (Normal landing) leg. The ordinate in fig. 6-1 to fig. 6-3 is the descent rate, the three graphs compare the descent rate ranges of the heavy landing class flight and the non-heavy landing class flight, for example, the descent rate of the heavy landing class flight is larger (the negative number of the ordinate indicates the direction, and the numerical value indicates the magnitude of the descent speed), in fig. 6-1 to fig. 6-3, the three descent rates (average, maximum, and minimum) of the heavy landing class flight are all larger than the descent rate of the non-heavy landing class, so that the parameter of the descent rate has higher importance for distinguishing the heavy landing class from the non-heavy landing class.
FIG. 7 shows a boxplot comparison of time _50ft _ td (time taken to ground at 50 ft), and from FIG. 7 it can be seen that the 50 feet to ground time for the heavy landing leg is significantly lower than the 50 feet to ground time for the non-heavy landing leg.
Fig. 8 shows the comparison result of the box line diagrams of min _ VRTG _ land (the minimum VRTG value in the landing phase), and it can be seen from the diagram that in the landing phase, the minimum VRTG value of the heavy landing leg is obviously lower than that of the non-heavy landing leg, and the fluctuation range of the minimum VRTG value of the heavy landing leg is larger, which indicates that the heavy landing leg is usually accompanied by larger VRTG fluctuation in the landing phase.
Fig. 9 shows a comparison result of the boxplot of max _ pitch _ land (the maximum pitch angle in the landing stage), and it can be seen from the plot that, in the landing stage, the maximum pitch angle in the heavy landing stage is significantly higher than that in the non-heavy landing stage, which indicates that there is a large attitude change in the heavy landing stage and there is no attitude stabilization in the non-heavy landing stage.
Fig. 10 shows a comparison of the boxplot of max _ pitch _ td (maximum pitch angle at the moment of touchdown), again with a significantly higher pitch angle for the heavy landing leg.
Fig. 11 gives a boxplot comparison of max _ pitch _50ft _ td (50ft to ground phase maximum pitch angle). Unlike before, here the maximum pitch angle of the heavy landing leg is not significantly higher than the non-heavy landing leg. The main difference is that the fluctuation range of the maximum pitch angle of the heavy landing leg is larger than that of the non-heavy landing leg, which indicates that the attitude change of the airplane in the heavy landing leg is more severe in the period. Thus, keeping the attitude of the aircraft stable at 50ft to the ground stage also helps to reduce the risk of heavy landing.
Fig. 12 shows the box plot comparison results of max _ abs _ roll _ td (maximum absolute roll angle at the time of grounding). As can be seen from fig. 12, the absolute value of the maximum roll angle of the heavy landing leg at the time of landing is significantly higher than that of the non-heavy landing leg, which indicates that the heavy landing leg generally has a relatively large roll attitude at the time of landing.
Fig. 13 shows the box plot comparison of max _ abs _ roll _50ft _ td (50ft to ground phase maximum absolute roll angle) with results similar to fig. 12.
The results of comparing box diagrams of the energy (kinetic energy + gravitational potential energy, unit: megajoule) of the aircraft at the time of grounding and 50ft height are shown in fig. 14-1 and fig. 14-2, respectively, and it can be seen that the average energy of the heavy landing leg is significantly higher than that of the non-heavy landing leg, which indicates that the energy has a certain influence.
15-1 and 15-2 show box plot comparison results of the ground speeds of the aircraft at the ground contact and 50ft altitude moments, respectively, and it can be seen that the average ground speed of the heavy landing leg is significantly higher than that of the non-heavy landing leg, indicating that the ground speed is too high, which increases the risk of heavy landing to some extent.
FIGS. 16-1 and 16-2 show box plot comparison results of maximum Ground Speed (GS) and maximum airspeed (IAS) during the landing phase. It can be seen that the maximum ground speed and airspeed of the heavy landing leg during the landing phase are significantly higher than those of the non-heavy landing leg.
FIG. 17 shows the comparison of the boxplot of max _ abs _ latg _ land (maximum lateral acceleration during landing) and shows that the heavy landing leg is significantly higher.
FIG. 18 shows a comparison of the min _ win _ lat _50ft _ td (50ft to ground phase minimum crosswind) box plots, and it can be seen that the minimum crosswind for the heavy landing leg is significantly greater, indicating that the heavy landing leg is generally experiencing a greater crosswind at 50ft to ground phase, further indicating that environmental factors (crosswinds) are one of the causes for the heavy landing event.
The risk assessment highlights that the risk degree of the heavy landing is assessed through comprehensive parameter information before landing for any flight data (including the heavy landing and the non-heavy landing), for example, the risk degree of the heavy landing of some flights is assessed through the heavy landing risk assessment when the 50ft time IVV is too large.
According to the statistical analysis result, the invention designs a heavy landing risk assessment model, which comprises two stages, namely a Measurement reference range establishing stage and a risk calculating stage, and the two stages are as follows:
(42) establishing a reference range of each Measurement index:
selecting Measurement indexes which are significant in distinguishing between heavy landing and non-heavy landing flights according to the box plot, and then calculating Measurement values of Measurement of non-heavy landing flights (the Measurement values of Measurement values are Measurement values of different time points or time periods of different parameters within 50 seconds from 30 seconds before grounding to 20 seconds after grounding, for example, max _ ivv _50ft _ td represents that the maximum descent rate value within the time period from 50 feet before grounding to grounding, and other parameters are obtained according to the Measurement data preprocessing part) within 90% of a range (meaning that 90% of the Measurement values are located within the range) of the Measurement indexes for the non-heavy landing flights on the basis of the indexes, wherein the Measurement values are used as reference ranges of the Measurement indexes.
And aiming at each selected Measurement index, giving a weight according to the capability of distinguishing heavy landing flight sections from non-heavy landing flight sections, wherein the value range of the weight is [ 0-1 ].
(43) Calculating a heavy landing risk value
And (3) aiming at one leg, firstly obtaining each measured value of the leg, then sequentially judging whether each measured value exceeds the reference range established in the step (42), if so, accumulating the Risk value of the measured value to the weight (namely multiplying the weight) corresponding to the measured index, and finally carrying out normalization processing on the Risk value to obtain the landing weight Risk value Risk of the leg.
Taking ivv (descent rate) at the time of grounding as an example, the reference range of ivv of non-re-landing is [ -200 to-400 ] (i.e. 90% of ivv is located in the range), for a given single leg, ivv of the grounding time of the leg is-500, i.e. -500 exceeds the range 100(-500- (-400) of the reference range), the absolute value is taken when calculating the size, i.e. the speed of 500 exceeds the speed 100 of the normal range (200 to 400),
the invention provides Measurement index weight related to heavy landing and a calculation Rule thereof, wherein the calculation Rule has three value types, namely Left, Right and Both, and the Rule indicates that the weight is accumulated in the heavy landing risk when the Measurement index of the leg deviates from the index reference range. For example, if the reference range of a Measurement index is [100-200 ], and the Measurement value of the Measurement index of the current leg is 50, the index of the current leg deviates to the left (i.e., is smaller than the lower limit of the reference range) from the index reference range.
The calculation formula (namely an evaluation model) of the heavy landing risk value is as follows:
Figure BDA0003420195790000151
where M represents the set of all Measurement indicators used to calculate risk (i.e., all selected Measurement indicators that distinguish significant between heavy and non-heavy landing legs), MiAnd the calculated risk value belongs to [0, 1] by normalizing the denominator by summing all weights when the condition in the bracket is established, namely the ith Measurement index, I {. is a symbolic function, and is 1 when the condition in the bracket is established, otherwise, I {. is 0]In the meantime. w is aiIs the weight of the ith Measurement index.
For the ith measurement index of the given navigation segment, judging whether the measurement value of the index deviates from the reference range from the left or the reference range from the right, wherein the left deviation is that the measurement value of the index is smaller than the lower limit part of the reference range, and the right deviation is that the measurement value of the index is larger than the upper limit part of the reference range. For example, the reference range is [100-]The measurement value of the index is 40, namely the left deviation reference range 60(100-40 is 60), the measurement value of the index is 230, namely the right deviation reference range 30(230-i"is to put the miThe index is matched to this rule and calculated accordingly.
"Left" means "only Left out" meets the rule, "Right out" does not meet the rule, "Right" otherwise; "booth" means that "either left or right run out" satisfies the rule. If the rule is "left out", then "match rule in the above formulai"indicates that I equals 1 when only left-biased, and I equals 0 when right-biased or within the reference range; if the rule is "Both", then "match rulei"indicates that I is equal to 1 for both left-hand and right-hand excursions, and is equal to 0 only within the reference range.
When the method is used, the remarkable Measurement indexes of the heavy landing and non-heavy landing flight legs are selected each time, then the reference range and the weight of each index are established, and finally the risk value is obtained according to a calculation formula.
No Measurement Weight Rule
1 avg_ivv_50ft_td 1 Left
2 max_ivv_50ft_td 0.7 Left
3 min_ivv_50ft_td 0.5 Left
4 time_50ft_td 0.6 Left
5 min_vrtg_land 0.3 Right
6 max_pitch_land 0.4 Right
7 max_pitch_td 0.3 Right
8 max_pitch_50ft_td 0.2 Both
9 max_abs_roll_td 0.3 Right
10 max_abs_roll_50ft_td 0.2 Right
11 energy_td 0.3 Right
12 energy_50ft 0.2 Right
13 gs_td 0.3 Right
14 gs_50ft 0.2 Right
15 max_gs_land 0.1 Right
16 max_ias_land 0.2 Right
17 gw_td 0.2 Right
18 ias_td 0.1 Right
19 max_abs_latg_land 0.2 Right
20 max_win_lat_50ft_td 0.1 Right
TABLE 6
The present invention counts 90% reference ranges of 49 Measurement indexes based on QAR data of 100 non-heavy landing legs, and the result is shown in table 7, where lower _ bound and upper _ bound respectively represent the left and right boundaries of the 90% reference ranges. 20 of table 6 are some of the relevant indicators listed to illustrate the weights of the different indicators, and 49 of table 7 are all indicators relevant to heavy landing, and are given as the boundary values of the reference range.
Figure BDA0003420195790000161
Figure BDA0003420195790000171
TABLE 7
According to the method, based on an evaluation model, the risk values of 52 heavy landing legs and 100 non-heavy landing legs are calculated, and the box plot comparison result is shown in fig. 19, so that the risk value (the average value is about 0.4) of the heavy landing leg is obviously higher than the risk value (the average value is about 0.05) of the non-heavy landing leg.
Fig. 20 further shows the relationship between the heavy landing risk value and the VRTG peak value through a scatter diagram, and as can be seen from fig. 20, overall, there is a certain positive correlation between the heavy landing risk value and the VRTG, and for the heavy landing leg, the relationship is more obvious, and for the non-heavy landing leg, the correlation is not very significant.
The invention also provides a flight landing quality monitoring and evaluating system, which comprises the following components:
the system comprises:
the device comprises an acquisition unit, a data acquisition unit and a data processing unit, wherein the acquisition unit is used for acquiring original QAR data of the airplane;
the extracting unit is connected with the collecting unit and used for extracting flight parameter measured values, namely Measurement data, from the original QAR data;
the preprocessing unit is respectively connected with the acquisition unit and the extraction unit and is used for preprocessing data;
the risk value calculation unit is connected with the preprocessing unit and used for constructing a heavy landing risk evaluation model to obtain a heavy landing risk value;
a visualization unit: and the risk value calculation unit is connected with the acquisition unit, the extraction unit, the preprocessing unit and the risk value calculation unit respectively and is used for visually displaying the data.
To facilitate the instrumentation of the above risk assessment model, the present invention designs this instrumentation system from three aspects.
The method comprises the steps of firstly, carrying out centralized storage management on model evaluation results of all flights (namely, risk values obtained by calculating each flight by the method), then automatically evaluating incremental flights (newly added flights) by using the model, and finally carrying out visual display on the landing quality of the flight from the whole dimensionality, wherein the model evaluation results can be used for measuring the landing quality of the airplane, namely the landing quality.
For the centralized management of the flight key data, the flight landing quality evaluation model needs some key data of flights to complete calculation, and in the face of tens of millions of flights per year, the method needs to acquire and standardize the management of the key data in time and support efficient query in time. For incremental flight assessment automation, the method is not only used for assessing historical flight segment data, but also for assessing the flight landing quality of newly added flights. When flight data is newly added, a model evaluation full flow of data acquisition, risk calculation and data storage needs to be automatically realized. Finally, for the visual display of the flight landing quality, valuable information such as risk change trend and the like is hidden in the mass flight data stored by the system, and the valuable information can be visualized.
The general framework of the system of the invention is shown in fig. 21, the data source is mainly provided by airguard (airguard is a decoding software, the most original QAR data needs to be converted into csv file after decoding through airguard so as to be convenient for data analysis), the backend database adopts MySQL to store some intermediate information of the system, the risk calculation is realized by Python, and the navigation segment risk value is obtained by calling Python flag interface and in the mode of HTTP request. The scheduling of the system is realized by XXLJOB, and the front end carries out the statistical analysis of the visual presentation risk of the page by Tableau. The system comprises an XXLJOB, a task scheduling platform and a task scheduling platform, wherein the XXLJOB is a lightweight distributed task scheduling platform and is used for scheduling and executing a timed task in the system to complete the functions of data acquisition, risk calculation and the like of incremental flights; OkHttp is an open source project for processing network requests, and is a lightweight framework of an android terminal; MyBatis is an excellent persistent layer framework that supports custom SQL, stored procedures, and advanced mapping, encapsulates native JDBC, and supports multi-style databases. The system is used for reading the data of the SqlServer database of the Airface and storing the data into the MySQL database; the FLASK is a practical lightweight Web application framework compiled by Python, is more flexible, portable, safe and easy to operate compared with other similar frameworks, a core algorithm for flight risk calculation is deployed on a WEB server built by the FLASK framework, and the system can finish the flight risk calculation by calling a data structure. MySQL is a relational database management system, and the management database stores data in different tables instead of putting all data in a large warehouse, so that the speed is increased and the flexibility is improved; tableau is mature BI software, and the tool provided by the invention can be used for visually displaying structured data such as flight risks and the like stored in MySQL.
The system of the present invention is developed and implemented in a front-end and back-end separated manner, as shown in fig. 22.
The back end of the system of the invention establishes a MySQL database which is used for storing flight information, measures, Reference, flight risk and other data; the method comprises the steps that a MyBatis framework based on JAVA is used for efficiently reading flight information, measures and other data stored in an SqlServer database of the airfast, and writing the data into a MySQL database of a system after structuring according to a configuration relation; calling a data interface by using an OkHttp technology, and storing a result in a MySQL database after finishing the landing quality calculation of the flight; the method is characterized in that an algorithm of a flight landing quality evaluation model is deployed on a WEB server based on a FLASK framework of Python, and the calculation is completed by system calling in a data interface mode.
The front end directly butts a MySQL database by using TABLEAU, and a visual chart is drawn by using structural data in the database; the trend change, the navigation department contrast and the display of key airports are realized by using a line graph, a box graph and the like, and meanwhile, the data aggregation and drilling are realized by reasonably using an instrument panel, and the effect is shown in fig. 22. The "screening" in fig. 22 is a screening of risk factors, that is, a screening of measurment indexes having a high degree of heavy landing correlation, and the screening is performed by preprocessing measurment and by drawing a box line graph (fig. 6 to fig. 18), and the risk values are calculated by using a risk evaluation model after the measurment indexes are screened.
The "risk factor display" in fig. 22 means that the measurement index is visualized through a box plot, and the correlation degree between the indicator and the heavy landing can be visually seen, and this part is a visualization part, namely, a tool action.
"calculating reference" in fig. 22 means calculating the reference range.
The "measure parameter configuration table" in fig. 22 is obtained by the measure data preprocessing section, and the configured measure parameters are obtained by configuring all QAR data within the original 50s to different measures indexes, for example, ivv index at the time of grounding and airspeed index at the height of 50 ft.
One embodiment of the invention is as follows:
[ examples ] A method for producing a compound
In the embodiment, one of the non-heavy landing legs (risk value: 0.3594) with the largest risk value is selected, and the vertical load (VRTG), the descent rate (IVV), the PITCH angle (PITCH) and the crosswind (CROSS _ WIND) are visualized, and the results are shown in fig. 24-1 to 24-4. It can be seen that although this leg did not trigger a heavy landing overrun event, its IVV curve has some similarity to the previously analyzed heavy landing leg IVV curve pattern, and the time it takes for this leg to go from 50ft to ground is short (about 5 seconds). Observing the pitch angle change, the attitude of the airplane is small when the airplane enters the height of 50ft, and the pitch angle is less than 1 degree. Through crosswind analysis, it can be seen that large crosswind (8-10 knots) always exists in the landing process of the flight segment. Therefore, through preliminary analysis, it can be judged that the segment risk is high due to the cross wind environmental factors.
According to the method, firstly, relevant Measurement indexes of heavy landing are extracted, then box line graphs of distribution ranges of heavy landing and non-heavy landing are compared for the indexes, indexes with obvious discrimination are selected, a Measurement reference range is established on the basis of the indexes, then the weight of each relevant index of heavy landing is designated according to the discrimination capability of the index, and the total risk value of heavy landing of the air leg is calculated by combining the value exceeding the index reference range in the previous stage. Actual results of experimental data show that the risk value of the heavy landing leg is significantly higher than that of the non-heavy landing leg, so that the effectiveness of the heavy landing risk assessment model provided by the invention is proved. In addition, in order to engineer the flight landing quality model, the invention designs a system, which performs centralized storage and management on the model evaluation results of all flights, automatically evaluates the incremental flight by using the model, and visually displays the flight landing quality from the whole dimension.
Finally, it should be noted that the above-mentioned technical solution is only one embodiment of the present invention, and it will be apparent to those skilled in the art that various modifications and variations can be easily made based on the application method and principle of the present invention disclosed, and the method is not limited to the above-mentioned specific embodiment of the present invention, so that the above-mentioned embodiment is only preferred, and not restrictive.

Claims (9)

1. A flight landing quality monitoring and evaluating method is characterized by comprising the following steps: the method utilizes original QAR data of the airplane to obtain a Measurement index related to the heavy landing and utilizes the Measurement index to obtain a heavy landing risk value.
2. The method of claim 1, wherein the method comprises: the method comprises the following steps:
(1) collecting original QAR data of an airplane;
(2) extracting flight parameter Measurement values, namely Measurement data, from original QAR data;
(3) preprocessing the data;
(4) and constructing a heavy landing risk evaluation model to obtain a heavy landing risk value.
3. The flying landing quality monitoring and evaluation method of claim 2, wherein: the operation of the step (2) comprises the following steps:
flight parameter measurements, namely, Measurement data, are extracted from the original QAR data using Measurement software.
4. The flying landing quality monitoring and evaluation method of claim 2, wherein: the operation of the step (3) comprises:
(31) preprocessing original QAR data;
(32) preprocessing the Measurement data.
5. The flying landing quality monitoring and evaluation method of claim 2, wherein: the operation of the step (4) comprises the following steps:
(41) extracting data related to heavy landing from the preprocessed Measurement data to serve as Measurement indexes, and drawing a box line graph of each Measurement index;
(42) establishing a reference range of each Measurement index, and determining the weight of each Measurement index;
(43) and obtaining a heavy landing risk value.
6. The flying landing quality monitoring and evaluation method of claim 5, wherein: the operation of drawing the box line graph of each Measurement index in the step (41) includes:
selecting a time period of 30 seconds forward and 20 seconds backward from the grounding time, and taking a total of 50 seconds as a landing interval;
for each flight segment, firstly extracting a change curve of a vertical load VRTG in a landing interval, then acquiring a peak value of the change curve, if the peak value is greater than a threshold value, judging that the flight segment is a heavy landing flight segment, and if the peak value is less than or equal to the threshold value, judging that the flight segment is a non-heavy landing flight segment;
and drawing a box line graph of each Measurement index, wherein the box line graph comprises a non-heavy landing leg and a heavy landing leg.
7. The flying landing quality monitoring and evaluation method of claim 6, wherein: the operation of step (42) comprises:
selecting a Measurement index which is obvious in distinguishing between heavy landing flight legs and non-heavy landing flight legs according to the box diagram;
distinguishing a remarkable Measurement index for each heavy landing flight leg and a non-heavy landing flight leg, obtaining a range of 90% of Measurement values of the Measurement index in a landing interval, and taking the range as a reference range of the Measurement index; and setting the weight of the Measurement index according to the capability of distinguishing the heavy landing leg from the non-heavy landing leg, wherein the value range of the weight is [ 0-1 ].
8. The flying landing quality monitoring and evaluation method of claim 7, wherein: the operation of said step (43) comprises:
and calculating to obtain a heavy landing Risk value Risk of one flight segment by using the following formula:
Figure FDA0003420195780000031
where M represents the set of all Measurement indicators used to calculate the heavy landing risk value, MiIs the ith Measurement index, I {. is a symbolic function, which represents: when the condition in brackets is established, I { } is 1, otherwise, I { } is 0, and wiIs the weight of the ith Measurement index;
the map rule indicates that the calculation rule is satisfied, which is as follows: the calculation rule includes three value types: left, Right and Both, where Left indicates that the measurement values are Left-shifted out of the reference range, Right indicates that the measurement values are Right-shifted out of the reference range, and Both indicates that the measurement values are Left-shifted out of the reference range or Right-shifted out of the reference range.
9. A flight landing quality monitoring and evaluation system is characterized in that: the system comprises:
the device comprises an acquisition unit, a data acquisition unit and a data processing unit, wherein the acquisition unit is used for acquiring original QAR data of the airplane;
the extracting unit is connected with the collecting unit and used for extracting flight parameter measured values, namely Measurement data, from the original QAR data;
the preprocessing unit is respectively connected with the acquisition unit and the extraction unit and is used for preprocessing data;
the risk value calculation unit is connected with the preprocessing unit and used for constructing a heavy landing risk evaluation model to obtain a heavy landing risk value;
a visualization unit: and the risk value calculation unit is connected with the acquisition unit, the extraction unit, the preprocessing unit and the risk value calculation unit respectively and is used for visually displaying the data.
CN202111559889.3A 2021-12-20 2021-12-20 Flight landing quality monitoring and evaluating method and system Pending CN114282792A (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202111559889.3A CN114282792A (en) 2021-12-20 2021-12-20 Flight landing quality monitoring and evaluating method and system

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202111559889.3A CN114282792A (en) 2021-12-20 2021-12-20 Flight landing quality monitoring and evaluating method and system

Publications (1)

Publication Number Publication Date
CN114282792A true CN114282792A (en) 2022-04-05

Family

ID=80873159

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202111559889.3A Pending CN114282792A (en) 2021-12-20 2021-12-20 Flight landing quality monitoring and evaluating method and system

Country Status (1)

Country Link
CN (1) CN114282792A (en)

Cited By (4)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115952694A (en) * 2023-03-13 2023-04-11 中国民用航空飞行学院 QAR data-based wind shear operation manipulation quality evaluation method in approach stage
CN116522771A (en) * 2023-04-21 2023-08-01 重庆大学 Attention mechanism-based bidirectional two-stage interpretable heavy landing prediction method
CN117194982A (en) * 2023-09-06 2023-12-08 中国民航科学技术研究院 Landing load risk early warning method and system for civil airliner and electronic equipment
CN117612415A (en) * 2024-01-24 2024-02-27 中国民用航空飞行学院 Landing safety assessment method and system based on flight data

Cited By (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN115952694A (en) * 2023-03-13 2023-04-11 中国民用航空飞行学院 QAR data-based wind shear operation manipulation quality evaluation method in approach stage
CN116522771A (en) * 2023-04-21 2023-08-01 重庆大学 Attention mechanism-based bidirectional two-stage interpretable heavy landing prediction method
CN116522771B (en) * 2023-04-21 2024-01-26 重庆大学 Attention mechanism-based bidirectional two-stage interpretable heavy landing prediction method
CN117194982A (en) * 2023-09-06 2023-12-08 中国民航科学技术研究院 Landing load risk early warning method and system for civil airliner and electronic equipment
CN117194982B (en) * 2023-09-06 2024-02-13 中国民航科学技术研究院 Landing load risk early warning method and system for civil airliner and electronic equipment
CN117612415A (en) * 2024-01-24 2024-02-27 中国民用航空飞行学院 Landing safety assessment method and system based on flight data

Similar Documents

Publication Publication Date Title
CN114282792A (en) Flight landing quality monitoring and evaluating method and system
US10867455B2 (en) Diagnostics, prognostics, and health management for vehicles using kinematic clusters, behavioral sensor data, and maintenance impact data
CN111259515B (en) Aircraft health management method and system
CN103247089A (en) System and method for real-time aircraft performance monitoring
CN109284896B (en) Human error mode influence analysis method
CN103761585A (en) Airport continuous transport dynamic matching, transport capacity monitoring and early warning and intelligent dispatching method
CN114004292B (en) Pilot flat-floating ejector rod behavior analysis method based on flight parameter data unsupervised clustering
CN112732687A (en) Aviation flight data visualization processing system and analysis method based on data cleaning
CN106651088A (en) Flight quality monitoring method based on temporal GIS
CN113298431A (en) Aviation QAR big data-based pilot competence portrayal method and system
Ackley et al. A supervised learning approach for safety event precursor identification in commercial aviation
CN106127407B (en) Airplane travel scoring method and system based on multi-sensor information fusion
CN114237110A (en) Multi-mode data driving-based general aviation fleet reliability monitoring and early warning system
CN113312424A (en) QAR data-based pilot flight skill portrait method and system
CN116415818B (en) Method and system for confirming risk points in aircraft approach stage based on clustering algorithm
CN106951700B (en) Approach stability assessment method based on energy management
CN115293225B (en) Method and device for analyzing causes of pilot flat-floating ejector rod
Li et al. Overview of application in data mining techniques to qar data ansys
Zixuan et al. Research on Influencing Factors of Fuel Flow Based on QAR Data
Janakiraman et al. Using ADOPT algorithm and operational data to discover precursors to aviation adverse events
CN114333428A (en) Airport landing aircraft interval management method, device and storage medium
Chen et al. Research on extraction of QAR key parameters during approach phase of civil aviation
CN116453377B (en) Method for carrying out flight phase division on airplane QAR data
Wu et al. Detecting and analyzing flight unstable approaches with QAR big data
CN117194982B (en) Landing load risk early warning method and system for civil airliner and electronic equipment

Legal Events

Date Code Title Description
PB01 Publication
PB01 Publication
SE01 Entry into force of request for substantive examination
SE01 Entry into force of request for substantive examination