CN116415818B - Method and system for confirming risk points in aircraft approach stage based on clustering algorithm - Google Patents

Method and system for confirming risk points in aircraft approach stage based on clustering algorithm Download PDF

Info

Publication number
CN116415818B
CN116415818B CN202310686392.0A CN202310686392A CN116415818B CN 116415818 B CN116415818 B CN 116415818B CN 202310686392 A CN202310686392 A CN 202310686392A CN 116415818 B CN116415818 B CN 116415818B
Authority
CN
China
Prior art keywords
data
risk
data points
airport
quadrant
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.)
Active
Application number
CN202310686392.0A
Other languages
Chinese (zh)
Other versions
CN116415818A (en
Inventor
韩静茹
焦洋
陈艳秋
孙华波
江书芳
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.)
China Academy of Civil Aviation Science and Technology
Original Assignee
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 China Academy of Civil Aviation Science and Technology filed Critical China Academy of Civil Aviation Science and Technology
Priority to CN202310686392.0A priority Critical patent/CN116415818B/en
Publication of CN116415818A publication Critical patent/CN116415818A/en
Application granted granted Critical
Publication of CN116415818B publication Critical patent/CN116415818B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Classifications

    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0635—Risk analysis of enterprise or organisation activities
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06F—ELECTRIC DIGITAL DATA PROCESSING
    • G06F18/00—Pattern recognition
    • G06F18/20—Analysing
    • G06F18/23—Clustering techniques
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06Q—INFORMATION AND COMMUNICATION TECHNOLOGY [ICT] SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES; SYSTEMS OR METHODS SPECIALLY ADAPTED FOR ADMINISTRATIVE, COMMERCIAL, FINANCIAL, MANAGERIAL OR SUPERVISORY PURPOSES, NOT OTHERWISE PROVIDED FOR
    • G06Q10/00—Administration; Management
    • G06Q10/06—Resources, workflows, human or project management; Enterprise or organisation planning; Enterprise or organisation modelling
    • G06Q10/063—Operations research, analysis or management
    • G06Q10/0639—Performance analysis of employees; Performance analysis of enterprise or organisation operations
    • G—PHYSICS
    • G06—COMPUTING; CALCULATING OR COUNTING
    • G06V—IMAGE OR VIDEO RECOGNITION OR UNDERSTANDING
    • G06V10/00—Arrangements for image or video recognition or understanding
    • G06V10/70—Arrangements for image or video recognition or understanding using pattern recognition or machine learning
    • G06V10/762—Arrangements for image or video recognition or understanding using pattern recognition or machine learning using clustering, e.g. of similar faces in social networks

Landscapes

  • Engineering & Computer Science (AREA)
  • Business, Economics & Management (AREA)
  • Human Resources & Organizations (AREA)
  • Theoretical Computer Science (AREA)
  • Strategic Management (AREA)
  • Physics & Mathematics (AREA)
  • Entrepreneurship & Innovation (AREA)
  • Economics (AREA)
  • General Physics & Mathematics (AREA)
  • Data Mining & Analysis (AREA)
  • Educational Administration (AREA)
  • Development Economics (AREA)
  • Computer Vision & Pattern Recognition (AREA)
  • Evolutionary Computation (AREA)
  • General Business, Economics & Management (AREA)
  • Quality & Reliability (AREA)
  • Operations Research (AREA)
  • Game Theory and Decision Science (AREA)
  • Artificial Intelligence (AREA)
  • Marketing (AREA)
  • Tourism & Hospitality (AREA)
  • General Health & Medical Sciences (AREA)
  • Life Sciences & Earth Sciences (AREA)
  • Computing Systems (AREA)
  • Medical Informatics (AREA)
  • Software Systems (AREA)
  • Multimedia (AREA)
  • Health & Medical Sciences (AREA)
  • Databases & Information Systems (AREA)
  • Bioinformatics & Cheminformatics (AREA)
  • Bioinformatics & Computational Biology (AREA)
  • Evolutionary Biology (AREA)
  • General Engineering & Computer Science (AREA)
  • Navigation (AREA)
  • Traffic Control Systems (AREA)

Abstract

The application belongs to the field of risk analysis, in particular relates to a method and a system for confirming risk points in an aircraft approach stage based on a clustering algorithm, and aims to solve the problem that hidden danger in a navigation stage is difficult to obtain by analyzing navigation risks only through navigation data of safety accidents. The application comprises the following steps: extracting a precursor index based on the historical safety event set to form a risk monitoring item; extracting the numerical value of a risk monitoring item in the historical safety event set and clustering to obtain a clustering distribution image; establishing a plane rectangular coordinate system based on the maximum safe flight data point of the maximum cluster in the cluster distribution image so as to divide the irregular data point; selecting a key airport based on the number of non-conventional data points; based on the historical safety data of the key airport, acquiring the preliminary risk cause of the key airport through a natural language processing algorithm. According to the application, by dividing the unconventional cluster data, possible risk causes can be detected, and the accuracy of confirming the risk points is improved.

Description

Method and system for confirming risk points in aircraft approach stage based on clustering algorithm
Technical Field
The application belongs to the field of risk analysis, and particularly relates to a method and a system for confirming risk points in an aircraft approach stage based on a clustering algorithm.
Background
Along with the change of the conception at the initial stage of construction brought by the operation of the airport, the middle and small airports are easy to have incomplete navigation facilities, configured personnel are difficult to meet the requirement of the change of the airport task, and part of airports belong to special airports such as shared airports or plateau/Gao Gaoyuan, and the operation environment is more complex compared with the conventional airports, so that the potential safety hazards of the middle and small airports and the special airports are more prominent.
Most of the existing risk analysis methods are based on triggered safety events, and analyze the reasons of event triggering, so that risk hidden dangers in the pilot operation process are found, and the analysis of the navigation planning of an airport is seldom performed. However, some risk hidden dangers do not trigger a safety event, but a pilot feels relatively high in operation difficulty in flight, and the pilot can stably land by adopting some unconventional means, namely, no accident occurs, but the operation of the pilot is difficult, and if the operation is improper, the potential safety hazard exists. The method aims to excavate the unconventional flight control, count which airports are relatively difficult to approach, find out the problem and adjust the navigation planning of the airports.
Disclosure of Invention
In order to solve the above problems in the prior art, namely the problem that hidden danger in a sailing stage is difficult to obtain by analyzing sailing risks only through sailing data of safety accidents, the application provides an aircraft approach stage risk point confirmation method based on a clustering algorithm, which comprises the following steps:
acquiring a historical security event set and QAR data;
extracting a precursor index based on the set of historical security events;
corresponding the precursor index to QAR data to form a risk monitoring item;
extracting the numerical value of a risk monitoring item in the historical safety event set, and converting the numerical value into scattered point data;
carrying out standardization processing on the scattered point data to obtain standardized scattered point data, and clustering the standardized scattered point data to obtain a clustering distribution image;
establishing a plane rectangular coordinate system based on the maximum safe flight data point of the maximum cluster in the cluster distribution image, and dividing the data point in the cluster distribution image into unconventional data points based on the rectangular coordinate system;
selecting a key airport based on the number of non-conventional data points;
based on historical safety data of a key airport, acquiring a preliminary risk cause of the key airport through a natural language processing algorithm, and acquiring the risk cause through QAR numerical analysis and QAR data comparison with other take-off and landing tasks of the same airport based on the preliminary risk cause.
In some preferred embodiments, the obtaining the risk cause based on the preliminary risk cause by QAR numerical analysis and QAR data comparison with other take-off and landing tasks of the same airport specifically includes:
and (3) correlating the flight number, the airplane number, the incident time and the landing airport information into QAR data, carrying out numerical analysis through the QAR data, and comparing the QAR data with QAR data of other landing tasks of the same airport to obtain risk causes.
In some preferred embodiments, the method further comprises the step of displaying the risk causes, in particular comprising:
and displaying an approach track through a map according to the approach parameters in the airport navigation database, displaying mark points on the map according to longitude and latitude information triggered by the QAR data monitoring items, and analyzing the approach track where the mark points are located to obtain the track risk cause.
In some preferred embodiments, the extracting the values of the risk monitoring items in the historical security event set is implemented by AGS decoding software.
In some preferred embodiments, the clustering of the normalized scatter data is performed by a DBScan algorithm.
In some preferred embodiments, the maximum safe flight data point is the coordinate value (V max ：Height max ) Maximum point, V max Indicating maximum airspeed, height, at landing gear down time max Indicating the maximum airspeed at the landing gear down time.
In some preferred embodiments, the dividing the data points in the cluster distribution image into irregular data points specifically includes:
in the rectangular coordinate system, the data points in the first quadrant, the second quadrant and the fourth quadrant are unconventional data points.
In some preferred embodiments, the selecting a key airport based on the number of non-regular data points specifically includes:
for each quadrant in the rectangular coordinate system, counting the number of the unconventional data points respectively, and sequencing at least the airports according to the number of the unconventional data points in the quadrant;
and selecting airports ranked within a preset range as key airports.
In some preferred embodiments, the natural language processing algorithms include a word segmentation algorithm and a TF-IDF algorithm.
In another aspect of the present application, an aircraft approach stage risk point confirmation system based on a clustering algorithm is provided, the system comprising: the system comprises a data acquisition module, a precursor index determination module, a monitoring item determination module, a clustering module, an irregular data point dividing module, a key airport selection module and a risk cause analysis module;
the data acquisition module is configured to acquire a historical security event set and QAR data;
the precursor indicator determination module is configured to extract precursor indicators based on the set of historical security events;
the monitoring item determining module is configured to correspond the precursor index to QAR data to form a risk monitoring item;
the clustering module is configured to extract the numerical value of the risk monitoring item in the historical safety event set and convert the numerical value into scattered point data; carrying out standardization processing on the scattered point data to obtain standardized scattered point data, and clustering the standardized scattered point data to obtain a clustering distribution image;
the unconventional data point dividing module is configured to establish a plane rectangular coordinate system based on the maximum safe flight data point of the maximum cluster in the cluster distribution image, and divide the data points in the cluster distribution image into unconventional data points based on the rectangular coordinate system;
the key airport selecting module is configured to select a key airport based on the number of unconventional data points;
the risk factor analysis module is configured to obtain preliminary risk factors of the key airport through a natural language processing algorithm based on historical safety data of the key airport, and obtain the risk factors through QAR data comparison with other take-off and landing tasks of the same airport based on the preliminary risk factors.
The application has the beneficial effects that:
(1) The application forms an algorithm suitable for near-stage risk analysis of flying based on a clustering idea, divides data with potential risk causes into different types by dividing the maximum cluster as a conventional cluster, and performs check analysis on the data of an airport more likely to have potential risk causes according to the number of unconventional cluster data, thereby improving the accuracy of identifying the risk causes and finding the potential risk causes.
Drawings
Other features, objects and advantages of the present application will become more apparent upon reading of the detailed description of non-limiting embodiments, made with reference to the accompanying drawings in which:
FIG. 1 is a flow chart of a clustering algorithm-based aircraft approach stage risk point confirmation method in an embodiment of the application;
FIG. 2 is a schematic flow diagram of airport risk analysis in an embodiment of the application;
FIG. 3 is a normalized scatter plot with landing gear down time altitude and airspeed as risk monitoring terms in an embodiment of the present application;
FIG. 4 is a cluster distribution image with landing gear down time altitude and airspeed as risk monitoring items in an embodiment of the present application;
FIG. 5 is a schematic illustration of the effect of dividing irregular data points with landing gear down time altitude and airspeed as risk monitoring items in an embodiment of the present application;
FIG. 6 is a chart of statistical ordering of airports with landing gear down time altitude and airspeed as the first quadrant of the risk monitoring item in an embodiment of the present application;
FIG. 7 is a chart of statistical ordering of airports in a second quadrant having landing gear down time altitude and airspeed as risk monitoring items in an embodiment of the present application;
FIG. 8 is a graph of statistical ordering of airports in the fourth quadrant of the landing gear down time altitude and airspeed as risk monitoring items in an embodiment of the present application;
fig. 9 is a schematic diagram of risk cause analysis for aircraft altitude, speed, and heading in an embodiment of the application.
Detailed Description
The application is described in further detail below with reference to the drawings and examples. It is to be understood that the specific embodiments described herein are merely illustrative of the application and are not limiting of the application. It should be noted that, for convenience of description, only the portions related to the present application are shown in the drawings.
It should be noted that, without conflict, the embodiments of the present application and features of the embodiments may be combined with each other. The application will be described in detail below with reference to the drawings in connection with embodiments.
In order to more clearly describe the method for confirming the risk points at the aircraft approach stage based on the clustering algorithm, each step in the embodiment of the application is described in detail below with reference to fig. 1 and 2.
The method for confirming the risk points at the aircraft approach stage based on the clustering algorithm in the first embodiment of the application comprises the steps S100-S800, wherein the detailed description of each step is as follows:
step S100, a historical security event set and QAR data are obtained.
The fast access recorder QAR (Quick Access Recorder) is an on-board flight data recorder designed to provide quick and convenient access to raw flight data the QAR receives input from a Flight Data Acquisition Unit (FDAU) to record over 2000 flight parameters.
In this embodiment, the AGS system extracts the control parameters and flight characteristics of the aircraft in the approach stage from the QAR data.
Step S200, extracting a precursor index based on the historical security event set.
The precursor index is process data extracted empirically.
And step S300, the precursor index is corresponding to the QAR data to form a risk monitoring item. The monitoring program can be independently set according to the risk monitoring items, the monitoring program comprises the steps of setting monitoring time, monitoring parameters and monitoring logic, the aircraft instrument parameters of each stage are extracted in a targeted mode, and the setting of the monitoring program is shown in the table 1:
table 1 monitor settings
Sequence number Monitoring time Monitoring parameters Monitoring logic
1 Turning moment Slope, speed and altitude Selecting aircraft magnetic heading parameterCounting HEAD, monitoring HEAD5 seconds ago and the current momentThe value of the change is about 20 degrees,the turning time is determined to be the turning time,setting the flag TURN1, otherwise TURN is 0.Selecting based on QAR dataThe flight phase being descent or descentIn the approach stage, collectOne second before TURN is 0Slope with time 1 in the first secondDegree, speed, altitude value.
2 Landing gear down time Speed, height Selecting landing gear control parametersNumber LDG_SEL, parameterIs the switching value, 0 tableShowing stow landing gear, table 1Landing gear is shown being lowered. Root of Chinese characterSelecting fly according to QAR dataThe travelling stage being falling or advancingNear stage, LDG_SEL frontOne second is 0, the current second is1. At this point, saveSpeed sum at the current timeHeight of the steel plate.
3 Flap change time Speed, height Selecting landing gear control parametersNumber CONF, parameter of oneIn an amount of 0 to 5Aircraft flap position. Root of Chinese characterSelecting fly according to QAR dataThe travelling stage being falling or advancingNear stage, CONF is formerSecond 0, current second 1At each moment, save the currentSpeed and altitude of time of day
4 Time of use of the speed reducing plate Speed, height
5 Heading alignment 5-edge moment 5 edge alignment distance Acquiring a flight landing airportRunway magnetic heading HEAD uLD_RUNWAY according to QARData, selecting flying stepsThe segments being descending or advancingSegment, magnetic heading parameterHEAD and HEAD_LD/uThe runray is compared to the running ray,if the difference is less than 3 degreesAnd last for 5 seconds, then considerThe aircraft heading is aligned to 5 sides.Accumulating by using ground speed integralGround speed GS of the aircraft untilThe aircraft is grounded. GS accumulationThe value is that the aircraft flies at 5 sidesLine length.
And step S400, extracting the numerical value of the risk monitoring item in the historical security event set, and converting the numerical value into scattered point data. In this embodiment, the QAR data of the navigation segment is derived and saved as a CSV file format by AGS decoding software implementation.
The values of risk monitoring items are extracted from historical security data for each aircraft at each airport as shown in table 2:
table 2 values of risk monitoring items for each airport
After the data are collected, the method further comprises the step of cleaning the data, and eliminating singular values caused by the quality of the data, such as airport or runway parameters being empty, speed and altitude values being-9999 and the like;
this embodiment studies the potential risk cause of a missed approach event as a function of approach phase, taking the altitude and airspeed at landing gear down time as an example.
And S500, carrying out standardization processing on the scattered point data to obtain standardized scattered point data, and clustering the standardized scattered point data to obtain a clustering distribution image. In this embodiment, implemented by the DBScan algorithm, eps is set to 0.05 and minPoints is set to 1.
The standardized scattered point data obtained by extracting and standardizing the height and the airspeed of the landing gear at the moment of lowering is shown in fig. 3, wherein the abscissa of the standardized scattered point data is the speed of the moment of lowering in fig. 3, the unit is a section, the ordinate is the height, and the unit is feet; as shown in fig. 4, the cluster distribution image obtained by clustering the standardized scatter data shows that each cluster is formed in fig. 4, and the cluster distribution image respectively represents different types of the landing gear down time and the landing gear down speed in the approach stage.
And S600, establishing a plane rectangular coordinate system based on the maximum safe flight data point of the maximum cluster in the cluster distribution image, and dividing the data points in the cluster distribution image into unconventional data points based on the rectangular coordinate system. In this embodiment, the maximum cluster represents the most present navigation conditions, and the present solution aims to find the hidden risk cause, and identify the data of the maximum cluster as the flight path of normal navigation without being affected by the potential safety hazard.
In this embodiment, the maximum safe flight data point is the coordinate value (V max ：Height max ) Maximum point, V max Indicating maximum airspeed, height, at landing gear down time max Indicating the maximum airspeed at the landing gear down time. The step of dividing the data points in the cluster distribution image into unconventional data points specifically comprises the following steps: in the rectangular coordinate system, the data points in the first quadrant, the second quadrant and the fourth quadrant are unconventional data points.
Taking fig. 5 as an example, taking the point at the top right corner of the largest cluster as the origin, and dividing all data points in the cluster distribution map into four quadrants; wherein the data points in the first quadrant represent the landing gear down time altitude and airspeed above the conventional value, the data points in the second quadrant represent the landing gear down time altitude above the conventional value and airspeed at the conventional value, and the data points in the fourth quadrant represent the landing gear down time altitude at the conventional value and airspeed above the conventional value.
Step S700, selecting a key airport based on the number of non-regular data points. The method specifically comprises the following steps: the selecting the key airport based on the number of the unconventional data points specifically comprises the following steps: for each quadrant in the rectangular coordinate system, counting the number of the unconventional data points respectively, and sequencing at least the airports according to the number of the unconventional data points in the quadrant; and selecting airports ranked within a preset range as key airports. As shown in fig. 6, 7 and 8, fig. 6 is the sorting of the airports according to the number of data points in the first quadrant, fig. 7 is the sorting of the airports according to the number of data points in the fourth quadrant, and fig. 8 is the sorting of the airports according to the number of data points in the second quadrant. The embodiment can select the airports with the ranking name of the top N airports as the key airports, or select the airports with the faults occurring in the number of data points as the key airports, or select the airports with the ranking name of the top N airports as the key airports according to the ratio of the data points to the total landing times of the airports.
The method for selecting the key airport further comprises the following steps:
respectively giving weight coefficients to the number of data points in the first quadrant, the number of data points in the second quadrant and the number of data points in the fourth quadrant, and calculating a potential risk score S:
wherein ,、/> and />Weight coefficient representing data point of first quadrant, weight coefficient of data point of second quadrant and weight coefficient of data point of fourth quadrant, respectively, +.>、/> and />The number of data points in the first quadrant, the number of data points in the second quadrant and the number of data points in the fourth quadrant of an airport are respectively;
counting the total number of the irregular data points appearing at each airport, selecting the airport with the ranking of at least the first 2N of the total number, calculating a first difference value of the number of the data points in the quadrant with the most irregular data points and the second number of the data points with the second most irregular data points of each selected airport, and a second difference value of the number of the data points with the most irregular data points and the number of the data points with the least irregular data points;
and counting the potential risk scores S, the first difference value and the second difference value of each airport, and setting the airports with the potential risk scores S ranked N times before, with the first difference value smaller than a preset gap threshold value or with the second difference value smaller than the preset gap threshold value as key airports.
In the embodiment, all QAR data are converted into regular data points and irregular data points, the types of the irregular data points are divided in a rectangular coordinate system mode, the data points in different quadrants indicate that different risk causes can exist, if the number of irregular data points in a certain quadrant of an airport is ranked at the front, the airport can have a certain category of risk causes, namely hidden danger, the airport is selected as a key airport, and the subsequent further analysis is guided; the data points are divided in different quadrants, so that the risk types can be divided into different categories, wherein the categories of the height and the airspeed at the landing gear falling time are larger than the conventional value need to be paid attention toGiving a larger value, but giving +.A higher than normal if only one of altitude or airspeed is higher than normal, if the risk potential is not negligible for a greater number of occurrences>Andwhen the risk score S ranks top, it is stated that there may be a more serious risk cause for the corresponding airport; there may be situations that the irregular data of each category of some airports are ranked and the risks of each category are not less, so that the ranking is performed according to the total number of the irregular data points, the first 2N airports with a larger range are selected for observation, the number of irregular flights occurring in the selected at least first 2N airports is worth noting, if the first difference value or the second difference value is too small, more than one risk cause may exist in the corresponding airports, and subsequent risk cause analysis is guided.
Step S800, acquiring a preliminary risk cause of a key airport through a natural language processing algorithm based on historical safety data of the key airport, and acquiring the risk cause through QAR numerical analysis and QAR data comparison with other take-off and landing tasks of the same airport based on the preliminary risk cause. The natural language processing algorithm comprises a word segmentation algorithm and a TF-IDF algorithm. According to the method, the possibility of risk causes can be checked one by one aiming at all monitoring items in a clustering mode, airport data with the same situation are arranged, various potential risk causes can be mutually verified, accuracy of risk point confirmation is improved, and accurate positions of locking risk causes in the whole navigation section are improved.
In this embodiment, the obtaining the risk cause based on the preliminary risk cause through QAR numerical analysis and QAR data comparison with other take-off and landing tasks of the same airport specifically includes:
and (3) correlating the flight number, the airplane number, the incident time and the landing airport information into QAR data, carrying out numerical analysis through the QAR data, and comparing the QAR data with QAR data of other landing tasks of the same airport to obtain risk causes. In this embodiment, the natural language processing algorithm may perform risk cause analysis on specific parameters of the whole navigation stage by means of keyword extraction, and table 3 is a keyword extraction table adopted in this embodiment;
TABLE 3 keyword extraction List
For the approach phase, further including extracting FAF point parameters, table 4 is an example of FAF point parameters;
TABLE 4 FAF Point parameters
The second embodiment of the present application proposes a step of displaying risk causes, specifically including:
the risk point confirmation method according to steps S100 to S800 acquires risk causes.
And displaying an approach track through a map according to the approach parameters in the airport navigation database, displaying mark points on the map according to longitude and latitude information triggered by the QAR data monitoring items, and analyzing the approach track where the mark points are located to obtain the track risk cause.
Taking 2017, 6 and 14 days as an example, an A320-220 aircraft executes a Qingdao-tin-free flight, 11:30, and in the visual approach process of a tin-free airport 03R runway, the aircraft cannot reach a stability standard after turning five sides due to late four turns, so that the aircraft flies at 368 feet of field height with accelerator, and then enters the visual approach program of the 03R runway again, and the method comprises the following steps: 40 normally fall to ground.
According to the requirement of establishing the stable form, the stable form is established before the aircraft reaches the height of 1000 feet, including landing gear down and flap extension, and part of special airports, such as A320-220 aircraft, execute the flying event described by Qingdao-tin-free flight; because the flight procedure is compact, the pilot workload is large, and unstable approach is easy to generate, so that the pilot flies. Thus, some pilots have used the approach of advancing the landing gear to establish a stable configuration earlier, which is an unusual maneuver for the subject study. Can be arranged into monitoring items.
The height and speed of landing gear at the descending stage of the model A320 are collected, and QAR data of the navigation stages are derived and stored as a CSV file format. The data is cleaned to remove singular values due to data quality, such as airport or runway parameters being empty, speed and altitude values being-9999, etc.
And taking the height and the speed of the landing gear at the moment of putting down in the tidied sample data as feature vectors, and carrying out visualization to generate scattered point data, wherein the abscissa is the speed, the unit is a section, the ordinate is the height, and the unit is feet.
After normalization, conventional data and non-conventional data are distinguished through a DBScan algorithm, four quadrants are counted respectively, and airports with the same situation are ranked.
In this embodiment, the first tin-free filling airport is selected as the key airport, and the QAR data and the security information thereof are extracted for analysis.
The near-flying event of the filling airport is obtained from the aeronautical security network, keyword extraction is carried out by using a word segmentation algorithm and TF-IDF, as shown in table 3, the keyword is found to represent 5-side, height, higher, stable and other word eyes, and further the keyword is interpreted as 5-side higher and unstable heading.
The aircraft altitude, speed and heading at the moment of alignment of the sides of the filling airport 5, namely parameters when the FAF (WX 903) point is analyzed, and the flight of the missed-flight event of the A320 model is found to have an outlier condition at the FAF point by analysis, as shown in fig. 9, in the circle of fig. 9, 1 flight capacity is larger, 3 flight headings are obviously higher, so that a stable form cannot be established, and derivative risks are generated.
Determining flight paths according to longitude and latitude recorded in an airline point basic information table, an approach database and an approach database, acquiring positions of landing airline points in the order of ESBAG- > SASAN- > WX912- > WX911- > WX903- > airport 03 according to the approach and approach database, marking and connecting lines on a map according to longitude and latitude information of each airline point, and recording positions of a flight landing line, a flight landing gear lowering time when a missed approach event occurs and a normal landing flight landing gear lowering time; compared with the airport aerial map, the landing gear releasing time of the flight landing gear when the flying-off event occurs is later, and the landing gear releasing time is matched with the high speed and the high altitude of the landing gear releasing time reflected by fig. 6. And determining that potential safety hazards exist when the landing gear is not properly put down.
Although the steps are described in the above-described sequential order in the above-described embodiments, it will be appreciated by those skilled in the art that in order to achieve the effects of the present embodiments, the steps need not be performed in such order, and may be performed simultaneously (in parallel) or in reverse order, and such simple variations are within the scope of the present application.
The third embodiment of the application provides an aircraft approach stage risk point confirmation system based on a clustering algorithm, which comprises a data acquisition module, a precursor index determination module, a monitoring item determination module, a clustering module, an unconventional data point division module, a key airport selection module and a risk cause analysis module;
the data acquisition module is configured to acquire a historical security event set and QAR data;
the precursor indicator determination module is configured to extract precursor indicators based on the set of historical security events;
the monitoring item determining module is configured to correspond the precursor index to QAR data to form a risk monitoring item;
the clustering module is configured to extract the numerical value of the risk monitoring item in the historical safety event set and convert the numerical value into scattered point data; carrying out standardization processing on the scattered point data to obtain standardized scattered point data, and clustering the standardized scattered point data to obtain a clustering distribution image;
the unconventional data point dividing module is configured to establish a plane rectangular coordinate system based on the maximum safe flight data point of the maximum cluster in the cluster distribution image, and divide the data points in the cluster distribution image into unconventional data points based on the rectangular coordinate system;
the key airport selecting module is configured to select a key airport based on the number of unconventional data points;
the risk cause analysis module is configured to obtain preliminary risk causes of the key airport through a natural language processing algorithm based on historical safety data of the key airport, and obtain the risk causes through QAR numerical analysis and QAR data comparison with other take-off and landing tasks of the same airport based on the preliminary risk causes.
It will be clear to those skilled in the art that, for convenience and brevity of description, the specific working process of the system described above and the related description may refer to the corresponding process in the foregoing method embodiment, which is not repeated here.
It should be noted that, in the aircraft approach stage risk point confirmation system based on the clustering algorithm provided in the foregoing embodiment, only the division of the foregoing functional modules is illustrated, in practical application, the foregoing functional allocation may be completed by different functional modules according to needs, that is, the modules or steps in the foregoing embodiment of the present application are further decomposed or combined, for example, the modules in the foregoing embodiment may be combined into one module, or may be further split into a plurality of sub-modules, so as to complete all or part of the functions described above. The names of the modules and steps related to the embodiments of the present application are merely for distinguishing the respective modules or steps, and are not to be construed as unduly limiting the present application.
It will be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working process of the storage device and the processing device described above and the related description may refer to the corresponding process in the foregoing method embodiment, which is not repeated herein.
Those of skill in the art will appreciate that the various illustrative modules, method steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both, and that the program(s) corresponding to the software modules, method steps, may be embodied in Random Access Memory (RAM), memory, read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, removable disk, CD-ROM, or any other form of storage medium known in the art. To clearly illustrate this interchangeability of electronic hardware and software, various illustrative components and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as electronic hardware or software depends upon the particular application and design constraints imposed on the solution. Those skilled in the art may implement the described functionality using different approaches for each particular application, but such implementation is not intended to be limiting.
The flowcharts and block diagrams in the figures illustrate the architecture, functionality, and operation of possible implementations of systems, methods and computer program products according to various embodiments of the present application. In this regard, each block in the flowchart or block diagrams may represent a module, segment, or portion of code, which comprises one or more executable instructions for implementing the specified logical function(s). It should also be noted that, in some alternative implementations, the functions noted in the block may occur out of the order noted in the figures. For example, two blocks shown in succession may, in fact, be executed substantially concurrently, or the blocks may sometimes be executed in the reverse order, depending upon the functionality involved. It will also be noted that each block of the block diagrams and/or flowchart illustration, and combinations of blocks in the block diagrams and/or flowchart illustration, can be implemented by special purpose hardware-based systems which perform the specified functions or acts, or combinations of special purpose hardware and computer instructions.
The terms "first," "second," and the like, are used for distinguishing between similar objects and not for describing a particular sequential or chronological order.
The terms "comprises," "comprising," or any other variation thereof, are intended to cover a non-exclusive inclusion, such that a process, method, article, or apparatus/apparatus that comprises a list of elements does not include only those elements but may include other elements not expressly listed or inherent to such process, method, article, or apparatus/apparatus.
Thus far, the technical solution of the present application has been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of protection of the present application is not limited to these specific embodiments. Equivalent modifications and substitutions for related technical features may be made by those skilled in the art without departing from the principles of the present application, and such modifications and substitutions will be within the scope of the present application.

Claims (7)

1. An aircraft approach stage risk point confirmation method based on a clustering algorithm is characterized by comprising the following steps:
acquiring a historical security event set and QAR data;
extracting a precursor index based on the set of historical security events;
corresponding the precursor index to QAR data to form a risk monitoring item;
extracting the numerical value of a risk monitoring item in the historical safety event set, and converting the numerical value into scattered point data;
carrying out standardization processing on the scattered point data to obtain standardized scattered point data, and clustering the standardized scattered point data to obtain a clustering distribution image;
establishing a plane rectangular coordinate system based on the maximum safe flight data point of the maximum cluster in the cluster distribution image, and dividing the data point in the cluster distribution image into unconventional data points based on the rectangular coordinate system;
the maximum safe flight data point is the coordinate value (V) in the maximum cluster in the cluster distribution image max ：Height max ) Maximum point, V max Indicating maximum airspeed, height, at landing gear down time max Representing the maximum height of the landing gear at the moment of landing gear down;
the step of dividing the data points in the cluster distribution image into unconventional data points specifically comprises the following steps:
in the rectangular coordinate system, the data points in the first quadrant, the second quadrant and the fourth quadrant are unconventional data points;
selecting a key airport based on the number of non-conventional data points; the method specifically comprises the following steps:
respectively giving weight coefficients to the number of data points in the first quadrant, the number of data points in the second quadrant and the number of data points in the fourth quadrant, and calculating a potential risk score S:
；
wherein ,、/> and />Weight coefficient representing data point of first quadrant, weight coefficient of data point of second quadrant and weight coefficient of data point of fourth quadrant, respectively, +.>、/> and />The number of data points in the first quadrant, the number of data points in the second quadrant and the number of data points in the fourth quadrant of an airport are respectively;
counting the total number of the irregular data points appearing at each airport, selecting the airport with the ranking of at least the first 2N of the total number, calculating a first difference value of the number of the data points in the quadrant with the most irregular data points and the second number of the data points with the second most irregular data points of each selected airport, and a second difference value of the number of the data points with the most irregular data points and the number of the data points with the least irregular data points;
counting the potential risk scores S, the first difference value and the second difference value of each airport, and setting the airports with the potential risk scores S ranked N times before, with the first difference value smaller than a preset gap threshold value or with the second difference value smaller than the preset gap threshold value as key airports;
based on historical safety data of a key airport, acquiring a preliminary risk cause of the key airport through a natural language processing algorithm, analyzing through QAR numerical values based on the preliminary risk cause, and comparing with QAR data of other take-off and landing tasks of the same airport to acquire the risk cause.
2. The method for confirming the near-stage risk points of the aircraft based on the clustering algorithm according to claim 1, wherein the obtaining the risk causes based on the preliminary risk causes through QAR numerical analysis and QAR data comparison with other take-off and landing tasks of the same airport specifically comprises the following steps:
and (3) correlating the flight number, the airplane number, the incident time and the landing airport information into QAR data, carrying out numerical analysis through the QAR data, and comparing the QAR data with QAR data of other landing tasks of the same airport to obtain risk causes.
3. The method for confirming the risk points at the aircraft approach stage based on the clustering algorithm according to claim 1, wherein the method further comprises the step of displaying the risk causes, and specifically comprises the following steps:
and displaying an approach track through a map according to the approach parameters in the airport navigation database, displaying mark points on the map according to longitude and latitude information triggered by the QAR data monitoring items, and analyzing the approach track where the mark points are located to obtain the track risk cause.
4. The method for confirming the risk points at the aircraft approach stage based on the clustering algorithm according to claim 1, wherein the extraction of the values of the risk monitoring items in the historical safety event set is realized by using AGS decoding software.
5. The method for confirming aircraft approach phase risk points based on clustering algorithm according to claim 1, wherein the clustering of the standardized scattered points is achieved through a DBScan algorithm.
6. The method for confirming aircraft approach phase risk points based on clustering algorithm according to claim 1, wherein the natural language processing algorithm comprises a word segmentation algorithm and a TF-IDF algorithm.
7. An aircraft approach stage risk point confirmation system based on a clustering algorithm, the system comprising: the system comprises a data acquisition module, a precursor index determination module, a monitoring item determination module, a clustering module, an irregular data point dividing module, a key airport selection module and a risk cause analysis module;
the data acquisition module is configured to acquire a historical security event set and QAR data;
the precursor indicator determination module is configured to extract precursor indicators based on the set of historical security events;
the monitoring item determining module is configured to correspond the precursor index to QAR data to form a risk monitoring item;
the clustering module is configured to extract the numerical value of the risk monitoring item in the historical safety event set and convert the numerical value into scattered point data; carrying out standardization processing on the scattered point data to obtain standardized scattered point data, and clustering the standardized scattered point data to obtain a clustering distribution image;
the unconventional data point dividing module is configured to establish a plane rectangular coordinate system based on the maximum safe flight data point of the maximum cluster in the cluster distribution image, and divide the data points in the cluster distribution image into unconventional data points based on the rectangular coordinate system;
the maximum safe flight data point is the coordinate value (V) in the maximum cluster in the cluster distribution image max ：Height max ) Maximum point, V max Indicating maximum airspeed, height, at landing gear down time max Representing the maximum height of the landing gear at the moment of landing gear down;
the step of dividing the data points in the cluster distribution image into unconventional data points specifically comprises the following steps:
in the rectangular coordinate system, the data points in the first quadrant, the second quadrant and the fourth quadrant are unconventional data points;
the key airport selecting module is configured to select a key airport based on the number of unconventional data points; the method specifically comprises the following steps:
respectively giving weight coefficients to the number of data points in the first quadrant, the number of data points in the second quadrant and the number of data points in the fourth quadrant, and calculating a potential risk score S:
；
wherein ,、/> and />Weight coefficient representing data point of first quadrant, weight coefficient of data point of second quadrant and weight coefficient of data point of fourth quadrant, respectively, +.>、/> and />The number of data points in the first quadrant, the number of data points in the second quadrant and the number of data points in the fourth quadrant of an airport are respectively;
counting the total number of the irregular data points appearing at each airport, selecting the airport with the ranking of at least the first 2N of the total number, calculating a first difference value of the number of the data points in the quadrant with the most irregular data points and the second number of the data points with the second most irregular data points of each selected airport, and a second difference value of the number of the data points with the most irregular data points and the number of the data points with the least irregular data points;
counting the potential risk scores S, the first difference value and the second difference value of each airport, and setting the airports with the potential risk scores S ranked N times before, with the first difference value smaller than a preset gap threshold value or with the second difference value smaller than the preset gap threshold value as key airports;
the risk cause analysis module is configured to obtain preliminary risk causes of the key airport through a natural language processing algorithm based on historical safety data of the key airport, and obtain the risk causes through QAR numerical analysis and QAR data comparison with other take-off and landing tasks of the same airport based on the preliminary risk causes.
CN202310686392.0A 2023-06-12 2023-06-12 Method and system for confirming risk points in aircraft approach stage based on clustering algorithm Active CN116415818B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN202310686392.0A CN116415818B (en) 2023-06-12 2023-06-12 Method and system for confirming risk points in aircraft approach stage based on clustering algorithm

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN202310686392.0A CN116415818B (en) 2023-06-12 2023-06-12 Method and system for confirming risk points in aircraft approach stage based on clustering algorithm

Publications (2)

Publication Number Publication Date
CN116415818A CN116415818A (en) 2023-07-11
CN116415818B true CN116415818B (en) 2023-09-12

Family

ID=87049628

Family Applications (1)

Application Number Title Priority Date Filing Date
CN202310686392.0A Active CN116415818B (en) 2023-06-12 2023-06-12 Method and system for confirming risk points in aircraft approach stage based on clustering algorithm

Country Status (1)

Country Link
CN (1) CN116415818B (en)

Families Citing this family (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN117114420B (en) * 2023-10-17 2024-01-05 南京启泰控股集团有限公司 Image recognition-based industrial and trade safety accident risk management and control system and method

Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107909106A (en) * 2017-11-14 2018-04-13 北京航空航天大学 A kind of detection method of aircraft flight environment
CN114819207A (en) * 2022-03-22 2022-07-29 中国民航科学技术研究院 Data evidence-based aircraft continuous airworthiness auxiliary management system and method
CN115293225A (en) * 2022-06-17 2022-11-04 重庆大学 Pilot flat drift ejector rod cause analysis method and device

Family Cites Families (1)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US11299288B2 (en) * 2019-03-20 2022-04-12 City University Of Hong Kong Method of presenting flight data of an aircraft and a graphical user interface for use with the same

Patent Citations (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN107909106A (en) * 2017-11-14 2018-04-13 北京航空航天大学 A kind of detection method of aircraft flight environment
CN114819207A (en) * 2022-03-22 2022-07-29 中国民航科学技术研究院 Data evidence-based aircraft continuous airworthiness auxiliary management system and method
CN115293225A (en) * 2022-06-17 2022-11-04 重庆大学 Pilot flat drift ejector rod cause analysis method and device

Non-Patent Citations (1)

* Cited by examiner, † Cited by third party
Title
"基于QAR数据聚类分析的航班异常检测研究";赵剑 等;《航空计算技术》;第48卷(第2期);52页右边栏-55页右边栏 *

Also Published As

Publication number Publication date
CN116415818A (en) 2023-07-11

Similar Documents

Publication Publication Date Title
EP3772030A1 (en) Systems and methods to utilize flight monitoring data
CN116415818B (en) Method and system for confirming risk points in aircraft approach stage based on clustering algorithm
CN110533095B (en) Flight risk behavior identification method based on improved random forest
US10696418B2 (en) Method and system for representation of flight events using icons within a graphical user interface
CN110390177B (en) Method and device for determining outlier flying object
Wang et al. The effect of psychological risk elements on pilot flight operational performance
CN102521901A (en) Method for automatically estimating and realizing flight quality of pilots
Hrastovec et al. Machine learning model for aircraft performances
Wang et al. Airspace risk management using surveillance track data: Stabilized approaches
Passarella et al. Data analysis of commercial aircraft landing on the runway airports in Indonesia
Wang et al. Data Mining on the Flight Quality of an Airline Based on QAR Big Data
LaBella et al. Optimized flight safety event detection in the national general aviation flight information database
Anderson et al. Analysis of flight delays in aviation system using different classification algorithms and feature selection methods
Zhang et al. Study on Evaluation of Airline Pilot’s Flight Violation Behaviors and Psychological Risk
GHALEBSAZ et al. A statistical analysis of the aircraft landing process
Wang et al. Predicting collisions between aircraft through spatiotemporal data-driven simulation of airport ground operations
Fang et al. An Aggregated Evaluation and Multi-dimensional Comparison Method of Flight Safety Based on QAR Data
Soares Flight Data Monitoring and its Application on Algorithms for Precursor Detection
Goblet Phase of flight identification in general aviation operations
Zhang et al. A Method of Analyzing Flight Exceedance Events Based on Association Rule and QAR Data
CN116453377B (en) Method for carrying out flight phase division on airplane QAR data
Cohen et al. Development of an aircraft performance risk assessment model
Hubbard et al. Using ADSB Data to Analyze the Potential for Nuisance in a Ground Collision Avoidance System
Kumar et al. Classification and Analysis of Go-Arounds in Commercial Aviation Using ADS-B Data. Aerospace 2021, 8, 291
Pavel et al. Utilizing Flight Data Monitoring for Near Miss Incident Analysis

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
GR01 Patent grant
GR01 Patent grant