CN113312424A - QAR data-based pilot flight skill portrait method and system - Google Patents
QAR data-based pilot flight skill portrait method and system Download PDFInfo
- Publication number
- CN113312424A CN113312424A CN202110721450.XA CN202110721450A CN113312424A CN 113312424 A CN113312424 A CN 113312424A CN 202110721450 A CN202110721450 A CN 202110721450A CN 113312424 A CN113312424 A CN 113312424A
- Authority
- CN
- China
- Prior art keywords
- pilot
- flight
- data
- index
- flight skill
- 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
Links
Images
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/26—Visual data mining; Browsing structured data
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F16/00—Information retrieval; Database structures therefor; File system structures therefor
- G06F16/20—Information retrieval; Database structures therefor; File system structures therefor of structured data, e.g. relational data
- G06F16/28—Databases characterised by their database models, e.g. relational or object models
- G06F16/284—Relational databases
- G06F16/285—Clustering or classification
- G06F16/287—Visualization; Browsing
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F18/00—Pattern recognition
- G06F18/20—Analysing
- G06F18/21—Design or setup of recognition systems or techniques; Extraction of features in feature space; Blind source separation
- G06F18/214—Generating training patterns; Bootstrap methods, e.g. bagging or boosting
-
- 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
- G06Q10/06398—Performance of employee with respect to a job function
Landscapes
- Engineering & Computer Science (AREA)
- Theoretical Computer Science (AREA)
- Business, Economics & Management (AREA)
- Databases & Information Systems (AREA)
- Human Resources & Organizations (AREA)
- Physics & Mathematics (AREA)
- Data Mining & Analysis (AREA)
- General Physics & Mathematics (AREA)
- General Engineering & Computer Science (AREA)
- Strategic Management (AREA)
- Development Economics (AREA)
- Educational Administration (AREA)
- Economics (AREA)
- Entrepreneurship & Innovation (AREA)
- Evolutionary Biology (AREA)
- Evolutionary Computation (AREA)
- Computer Vision & Pattern Recognition (AREA)
- Bioinformatics & Computational Biology (AREA)
- Bioinformatics & Cheminformatics (AREA)
- Game Theory and Decision Science (AREA)
- Artificial Intelligence (AREA)
- Marketing (AREA)
- Operations Research (AREA)
- Quality & Reliability (AREA)
- Tourism & Hospitality (AREA)
- General Business, Economics & Management (AREA)
- Life Sciences & Earth Sciences (AREA)
- Management, Administration, Business Operations System, And Electronic Commerce (AREA)
Abstract
The invention relates to a pilot flight skill portrait method and a pilot flight skill portrait system based on QAR data, which comprises the following steps: acquiring multiple flight task data of a set machine type in a set period; determining effective data according to the multiple flight task data; screening the effective data to construct a pilot flight skill portrait index system; grading the flight skill dimension of the pilot by utilizing a grey correlation method according to the pilot flight skill portrait index system, and constructing a portrait model of the flight skill; and constructing a radar chart of the multi-dimensional flight skill portrait of the pilot according to the portrait model. The invention can improve the comprehensiveness and pertinence of the portrait of the pilot.
Description
Technical Field
The invention relates to the technical field of flight quality monitoring management, in particular to a pilot flight skill portrait method and system based on QAR data.
Background
According to civil aviation accident statistics, human factors are considered to be the most main reason in current flight accidents. The pilot acts as a direct operator of the aircraft, and the operation behavior and skill level of the pilot directly influence the operation safety of the aircraft, and if the pilot fails to operate, the pilot causes inestimable loss. Therefore, the flight skill and the operation characteristics of the pilot are comprehensively and pertinently evaluated, the operation behaviors which do not meet the requirements or have risk tendency are identified in time, and corresponding training and management schemes are matched, so that the method is an important way for avoiding and reducing flight accidents and unsafe events.
In the aspect of evaluating the flight skill and flight operation characteristics of pilots, researchers at home and abroad have carried out many researches, and the current researches are mainly based on QAR (Quick Access Recorder) data and describe the flight capability of pilots by analyzing various operation indexes of the pilots in the flight process. However, the existing research methods still have some defects. On the one hand, the flight skill dimension is constructed on one side. When the flight skill dimension is established, the conventional research method mainly adopts a mode of manually screening flight key point indexes, takes a single index as the dimension for measuring a certain flight skill, and evaluates the flight capability by determining an overrun event. However, this "point-and-cover" analysis method is clearly not able to adequately measure the overall level of flight skills. In fact, a flight skill is generally related to a plurality of indexes, such as the control capability of the lift rate of a pilot during the flight, the lift rate index of the airplane at different heights needs to be comprehensively considered, and even other indexes related to the lift rate need to be further mined and combined. On the other hand, the flight skill portrait model is too simple. The existing method for the capability sketch of the pilot mainly takes single evaluation methods such as expert scoring, descriptive statistics and the like as main points, the flight skill of the pilot is roughly graded mainly by analyzing whether the pilot has an overrun event, overrun event frequency and overrun degree, and few researches can establish a set of more accurate capability grading model aiming at different flight skill standards, so that the constructed capability sketch is macroscopic and simple, and the targeted analysis of the pilot on the aspect of operation technical characteristics is lacked.
Disclosure of Invention
The invention aims to provide a pilot flight skill portrait method and system based on QAR data so as to improve comprehensiveness and pertinence of the portrait of a pilot.
In order to achieve the purpose, the invention provides the following scheme:
a pilot flight skill representation method based on QAR data comprises the following steps:
acquiring multiple flight task data of a set machine type in a set period;
determining effective data according to the multiple flight task data;
screening the effective data to construct a pilot flight skill portrait index system;
grading the flight skill dimension of the pilot by utilizing a grey correlation method according to the pilot flight skill portrait index system, and constructing a portrait model of the flight skill;
and constructing a radar chart of the multi-dimensional flight skill portrait of the pilot according to the portrait model.
Optionally, the determining effective data according to the multiple flight mission data specifically includes:
preprocessing the flight task data to determine effective data; the preprocessing includes outlier processing and missing value processing.
Optionally, the screening of the effective data to construct a pilot flight skill portrait index system specifically includes:
screening the effective data according to the variance of the effective data to determine an effective evaluation data set; the valid evaluation data set comprises an original index;
standardizing the effective evaluation data set to obtain a standardized data set;
clustering the indexes in the standardized data set by using a principal component analysis method to determine new indexes;
screening the new indexes according to a linear transformation relation to determine flight skill dimensions; the linear transformation relation is a linear transformation relation between the original index and the new index;
and constructing a pilot flight skill image index system according to the effective evaluation data set and the flight skill dimension.
Optionally, the clustering the indexes in the normalized data set by using a principal component analysis method to determine new indexes specifically includes:
calculating index correlation according to the standardized data set to obtain a correlation coefficient matrix;
determining the eigenvalue of the correlation coefficient matrix by using a Jacobi method;
determining a unit feature vector and an information contribution rate according to the feature value;
and clustering the indexes in the standardized data set according to the unit feature vector and the information contribution rate to determine a new index.
Optionally, the scoring is performed on the flight skill dimension of the pilot by using a gray correlation method according to the pilot flight skill portrait index system, and the building of the portrait model of the flight skill specifically includes:
determining the ideal value of the original index corresponding to the ideal value of the new index by utilizing the linear transformation relation according to the ideal value of the original index;
determining an absolute difference value according to the ideal value of the new index; the absolute difference is a difference between an ideal value of the new index and an actual value of the new index;
determining a correlation coefficient according to the absolute difference value; the correlation coefficient is a correlation coefficient between an ideal value of the new index and an actual value of the new index;
and determining the flight skill dimension score of the pilot according to the correlation coefficient to obtain an image model of the flight skill.
Optionally, the building of the multidimensional flight skill portrait radar chart of the pilot according to the portrait model specifically includes:
and constructing a multidimensional flight skill portrait radar chart of the pilot by utilizing a visual application program package according to the pilot flight skill dimension score in the portrait model.
A pilot flight skill representation system based on QAR data, comprising:
the acquisition module is used for acquiring multiple flight mission data of a set machine type in a set period;
the effective data determining module is used for determining effective data according to the multiple flight mission data;
the effective data screening module is used for screening the effective data and constructing a pilot flight skill portrait index system;
the scoring module is used for scoring the flight skill dimension of the pilot by utilizing a grey correlation degree method according to the pilot flight skill portrait index system and constructing a portrait model of the flight skill;
and the multidimensional flight skill portrait radar chart building module is used for building a multidimensional flight skill portrait radar chart of the pilot according to the portrait model.
Optionally, the valid data screening module specifically includes:
the effective data screening unit is used for screening the effective data according to the variance of the effective data to determine an effective evaluation data set; the valid evaluation data set comprises an original index;
the standardization processing unit is used for carrying out standardization processing on the effective evaluation data set to obtain a standardized data set;
a new index determination unit for clustering the indexes in the normalized data set by using a principal component analysis method to determine new indexes;
the flight skill dimension determining unit is used for screening the new indexes according to a linear transformation relation and determining the flight skill dimension; the linear transformation relation is a linear transformation relation between the original index and the new index;
and the pilot flight skill portrait index system construction unit is used for constructing a pilot flight skill portrait index system according to the effective evaluation data set and the flight skill dimension.
Optionally, the new indicator determining unit specifically includes:
a correlation coefficient matrix determining subunit, configured to calculate index correlation according to the normalized data set, so as to obtain a correlation coefficient matrix;
an eigenvalue determination subunit, configured to determine an eigenvalue of the correlation coefficient matrix by using a jacobian method;
the unit characteristic vector and information contribution rate determining subunit is used for determining the unit characteristic vector and the information contribution rate according to the characteristic value;
and the new index determining subunit is used for clustering the indexes in the normalized data set according to the unit feature vector and the information contribution rate to determine a new index.
Optionally, the scoring module specifically includes:
the ideal value determining unit is used for determining the ideal value of the original index corresponding to the ideal value of the new index according to the ideal value of the original index by utilizing the linear transformation relation;
an absolute difference value determination unit for determining an absolute difference value from the ideal value of the new index; the absolute difference is a difference between an ideal value of the new index and an actual value of the new index;
a correlation coefficient determining unit, configured to determine a correlation coefficient according to the absolute difference; the correlation coefficient is a correlation coefficient between an ideal value of the new index and an actual value of the new index;
and the image model determining unit is used for determining the flight skill dimension score of the pilot according to the correlation coefficient to obtain an image model of the flight skill.
According to the specific embodiment provided by the invention, the invention discloses the following technical effects:
according to the method and the system for the pilot flight skill portrait based on the QAR data, provided by the invention, the flight task data is screened through the thought of big data, so that a pilot flight skill portrait index system is constructed, in addition, the flight skill dimensionality of the pilot is graded through a grey correlation method, a portrait model of the flight skill is constructed, the trend level of the individual dimensionality capacity of the pilot can be determined, and the comprehensiveness and pertinence of the portrait of the pilot are improved.
Drawings
In order to more clearly illustrate the embodiments of the present invention or the technical solutions in the prior art, the drawings needed to be used in the embodiments will be briefly described below, and it is obvious that the drawings in the following description are only some embodiments of the present invention, and it is obvious for those skilled in the art to obtain other drawings without inventive exercise.
FIG. 1 is a flow chart of a pilot flight skill representation method of the present invention based on QAR data;
FIG. 2 is a simplified flow chart of a pilot flight skill representation method of the present invention based on QAR data;
FIG. 3 is a six-dimensional radar plot of a flight skill representation during pilot landing;
FIG. 4 is a schematic view of a pilot flight skill representation system based on QAR data in accordance with the present invention.
Detailed Description
The technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are only a part of the embodiments of the present invention, and not all of the embodiments. All other embodiments, which can be derived by a person skilled in the art from the embodiments given herein without making any creative effort, shall fall within the protection scope of the present invention.
The invention aims to provide a pilot flight skill portrait method and system based on QAR data so as to improve comprehensiveness and pertinence of the portrait of a pilot.
In order to make the aforementioned objects, features and advantages of the present invention comprehensible, embodiments accompanied with figures are described in further detail below.
As shown in FIGS. 1-2, the present invention provides a pilot flight skill representation method based on QAR data, comprising:
step 101: and acquiring multiple flight task data of the set model in a set period.
Step 102: and determining effective data according to the multiple flight task data. Wherein, step 102 specifically includes: preprocessing the flight task data to determine effective data; the preprocessing includes outlier processing and missing value processing. The data of multiple flight tasks of a certain model in a certain period are collected through an airline company QAR system, an abnormal value with large deviation in a QAR data set is deleted by adopting a quartile method, and data containing a missing value are removed to obtain an effective data set.
Step 103: and screening the effective data to construct a pilot flight skill portrait index system.
The indexes with evaluation value on the flight skill are screened out by combining the characteristics of expert experience and index values, the dimensionality reduction and clustering are carried out on the multi-dimensional QAR data indexes according to the correlation among the indexes, the representative dimensionality capable of reflecting the flight skill of the pilot is further determined according to expert opinions, and support is provided for the next portrait work.
Step 103 specifically includes:
screening the effective data according to the variance of the effective data to determine an effective evaluation data set; the valid evaluation data set includes an original index.
And carrying out standardization processing on the effective evaluation data set to obtain a standardized data set.
Clustering the indexes in the standardized data set by using a principal component analysis method to determine new indexes; and clustering the indexes in the standardized data set by using a principal component analysis method to determine a new index, specifically calculating the index correlation according to the standardized data set to obtain a correlation coefficient matrix. And determining the eigenvalue of the correlation coefficient matrix by using a Jacobi method. And determining a unit feature vector and an information contribution rate according to the feature value. And clustering the indexes in the standardized data set according to the unit feature vector and the information contribution rate to determine a new index.
Screening the new indexes according to a linear transformation relation to determine flight skill dimensions; the linear transformation relation is a linear transformation relation between the original index and the new index.
And constructing a pilot flight skill image index system according to the effective evaluation data set and the flight skill dimension.
Step 104: and grading the flight skill dimension of the pilot by utilizing a grey correlation method according to the pilot flight skill portrait index system, and constructing a portrait model of the flight skill.
Wherein, step 104 specifically includes:
and determining the ideal value of the original index corresponding to the ideal value of the new index by utilizing the linear transformation relation according to the ideal value of the original index.
Determining an absolute difference value according to the ideal value of the new index; the absolute difference is a difference between an ideal value of the new index and an actual value of the new index.
Determining a correlation coefficient according to the absolute difference value; the correlation coefficient is a correlation coefficient between an ideal value of the new index and an actual value of the new index.
And determining the flight skill dimension score of the pilot according to the correlation coefficient to obtain an image model of the flight skill.
Step 105: and constructing a radar chart of the multi-dimensional flight skill portrait of the pilot according to the portrait model.
Wherein, step 105 specifically includes:
and constructing a multidimensional flight skill portrait radar chart of the pilot by utilizing a visual application program package according to the pilot flight skill dimension score in the portrait model.
As shown in FIG. 2, the present invention also provides a specific flow of the pilot flight skill representation method based on QAR data in practical application, comprising the following steps:
step A: preprocessing data; the data of multiple flight tasks of a certain model in a certain period are collected through an airline QAR system, abnormal values and missing values of the data are processed, and effective data are obtained. Specifically, an abnormal value with large deviation in a QAR data set is deleted by adopting a quartile method, and data containing a missing value is removed to obtain effective data.
And B: constructing a pilot flight skill portrait index system; the method is characterized in that the method combines the characteristics of expert experience and index values to screen effective data, screens out indexes with evaluation value on flight skills, reduces and clusters the multidimensional QAR data indexes according to the correlation among the indexes, further determines representative dimensions capable of reflecting the flight skills of pilots according to expert opinions, and provides support for the next portrait work. The method comprises the following steps:
step B1: screening indexes; and (4) deleting the index with low flight skill evaluation value by combining the expert experience and the characteristics of the index value to obtain a final effective data set.
The method comprises the following specific steps: firstly, deleting indexes (invalid indexes) with the same numerical values; ② sampling N by airline expert1Task with good flight performance and N2The tasks with poor performance are respectively used as a training set 1 and a training set 2, if the variance of a certain index in the training set 1 is greater than the variance of the index in the training set 2 and all data sets, the result that the indexes have a large degree of change in the training set 1 but do not affect the good performance of the whole training set 1 is shown, and the indexes are considered to have a small evaluation value on the flight skill and are deleted. Finally obtaining an effective evaluation data set X ═ X (X) containing n samples and m evaluation indexesij)n×mWherein x isijAnd representing the specific numerical value of the j index under the ith task for each element in the original data set.
Step B2: determining a flight skill dimension; based on the principal component analysis method, clustering of flight operation indexes is achieved by mining the correlation among the indexes, and the portrait dimension capable of representing specific flight skills is determined by further combining the clustering result and expert experience.
The method comprises the following specific steps: (X) for the valid evaluation dataset X ═ Xij)n×mPerforming a normalization process, i.e.
(i 1,2, …, n; j 1, 2.. multidot.m), wherein,
is the average value of the j-th index,
sjis the standard deviation of the jth index, n and m respectively represent the number of samples and the number of evaluation indexes, x'ijThe normalized data set X' is obtained for the elements in the normalized data set (i ═ 1, 2.. times.n) (j ═ 1, 2.. times.m).ij)n×m(ii) a ② calculating the correlation between indexes to obtain correlation coefficient matrix R ═ Rij)m×mWherein
rii＝1，rij＝rji，(i,j＝1,2,...,m)，rijAnd rjiAll represent correlation coefficients, x'ki、x’kjAre all elements in the normalized dataset; calculating a correlation coefficient matrix R ═ R (R) by adopting a Jacobi (Jacobi) methodij)m×mIs arranged in order of magnitude, i.e. λ1≥λ2≥…≥λmIs greater than or equal to 0, and respectively calculates the corresponding characteristic value lambdaiUnit feature vector a ofiWherein a isi＝(ai1,ai2,...,aim)T(ii) a Fourthly, calculating the characteristic value lambdaiInformation contribution rate of
And according to cumulative contribution rate
Determining the first k principal components, and using each principal component as a new index YiAnd Y isiThe linear transformation relationship with the original index can be expressed as follows:
wherein X'j＝(x’1j,x’2j,...,x’nj)，(j＝1,2,…,m)，X’jData normalized for the j-th index, aijIs a vector aiThe j-th component of (i 1, 2.., k) (j 1, 2.., m). Y iskThe actual value of the new index is obtained by the conversion relation of the original index through the formula (1). Fifthly, combining the coefficient and expert opinions of the formula (1), further screening p indexes Y capable of representing the specific flight skills of the pilot from the k new indexes1*,Y2*,…,Yp*As p dimensions of the image. The flight skill dimensionality determined by QAR data in the landing process mainly comprises the following steps: grade control (associated with the height of the original index containing the character "ROLL"), vertical overload control (associated with the height of the original index containing the character "VRTG"), DRIFT control (associated with the height of the original index containing the character "DRIFT"), ground speed control (associated with the height of the original index containing the character "GS"), rate-of-rise control (associated with the height of the original index containing the character "IVV"), and PITCH control (associated with the height of the original index containing the character "PITCH").
And C: establishing a flight skill portrait model of a pilot; and B, based on the pilot flight skill portrait index system established in the step B, scoring each flight skill dimension of the pilot by adopting a grey correlation method, and constructing a portrait model capable of reflecting flight skills.
The specific meanings are as follows: and B, based on the pilot flight skill portrait index system established in the step B, scoring each flight skill dimension of the pilot by adopting a grey correlation degree method, wherein the portrait index system is composed of the indexes screened out in the step B1 and the flight skill dimensions established in the step B2. The method specifically comprises the following steps: making ideal values X 'according to original indexes'0＝(x’01,x’02,…,x’0m) And formula (1) for calculating the specific value Y of the ideal value corresponding to the new index0＝(y01*,y02*,…,y0p*) Wherein y is0j＝aj1x’01+aj2x’02+…+ajmx’0m，x’0jIs the ideal value of the jth original index, where (j ═ 1,2, …, m). Y is0The ideal value of the new index is obtained by the ideal value of the original index through the conversion relation of the formula (1); ② calculating absolute difference value deltaij＝|y0j-yijL, where yijIs a specific numerical value of the jth new index of the ith flight mission, and calculates the maximum difference and the minimum difference of two stages,
wherein Max is the maximum difference of two levels, and Min is the minimum difference of two levels. Calculating yijCorrelation coefficient with ideal value under new index
Taking the obtained value as the score of the flight mission in each dimension, wherein rho epsilon (0,1) is a resolution coefficient, and rho is generally taken to be 0.5; fourthly, taking continuous u times of tasks of each pilot as a flight cycle, and taking the average value of all dimension scores of the u times of tasks as all dimension capability scores of the pilot f, namely
The higher the score, the higher the pilot's ability level in that dimension.
Step D: constructing a radar map of the flight skill portrait of the pilot; and C, constructing a multidimensional flight skill portrait radar chart of the pilot through a plotly (visual application program package) tool in R (computer program design) software based on the scoring values of the flight skill dimensions of the pilot, which are obtained through calculation in the step C.
As shown in FIG. 3, the embodiment can present the pilot flight skill representation in the form of six-dimensional radar graph according to the flight skill dimension determined by QAR data during landing, and the capability score of the pilot changes along with the change of certain time sequence dynamic operation data of the pilot. Through transverse comparison and analysis of the flight skill radar map, the capability tendency level and weak links of the pilot can be found, and then the training scheme is pertinently adjusted by taking data as a theoretical basis; through the longitudinal comparison and analysis of the flight skill radar map, the whole flight fleet can be comprehensively evaluated, and theoretical support is provided for management and training of an airline company.
As shown in FIG. 4, the present invention provides a pilot flight skill representation system based on QAR data, comprising:
the acquiring module 401 is configured to acquire multiple flight mission data of a set model in a set period.
And an effective data determining module 402, configured to determine effective data according to the multiple flight mission data.
And the effective data screening module 403 is configured to screen the effective data to construct a pilot flight skill portrait index system.
And the scoring module 404 is used for scoring the flight skill dimension of the pilot by using a grey correlation method according to the pilot flight skill portrait index system and constructing a portrait model of the flight skill.
And a multidimensional flight skill portrait radar chart building module 405, configured to build a multidimensional flight skill portrait radar chart of the pilot according to the portrait model.
In practical applications, the valid data screening module 403 specifically includes:
the effective data screening unit is used for screening the effective data according to the variance of the effective data to determine an effective evaluation data set; the valid evaluation data set includes an original index.
And the standardization processing unit is used for carrying out standardization processing on the effective evaluation data set to obtain a standardized data set.
And a new index determination unit for clustering the indexes in the normalized data set by using a principal component analysis method to determine new indexes. The new index determining unit specifically includes: a correlation coefficient matrix determining subunit, configured to calculate index correlation according to the normalized data set, so as to obtain a correlation coefficient matrix; and the eigenvalue determination subunit is used for determining the eigenvalue of the correlation coefficient matrix by using a Jacobian method. And the unit characteristic vector and information contribution rate determining subunit is used for determining the unit characteristic vector and the information contribution rate according to the characteristic value. And the new index determining subunit is used for clustering the indexes in the normalized data set according to the unit feature vector and the information contribution rate to determine a new index.
The flight skill dimension determining unit is used for screening the new indexes according to a linear transformation relation and determining the flight skill dimension; the linear transformation relation is a linear transformation relation between the original index and the new index.
And the pilot flight skill portrait index system construction unit is used for constructing a pilot flight skill portrait index system according to the effective evaluation data set and the flight skill dimension.
In practical applications, the scoring module 404 specifically includes:
and the ideal value determining unit is used for determining the ideal value of the original index corresponding to the ideal value of the new index by utilizing the linear transformation relation according to the ideal value of the original index.
An absolute difference value determination unit for determining an absolute difference value from the ideal value of the new index; the absolute difference is a difference between an ideal value of the new index and an actual value of the new index.
A correlation coefficient determining unit, configured to determine a correlation coefficient according to the absolute difference; the correlation coefficient is a correlation coefficient between an ideal value of the new index and an actual value of the new index.
And the image model determining unit is used for determining the flight skill dimension score of the pilot according to the correlation coefficient to obtain an image model of the flight skill.
The invention provides a pilot flight skill portrait method based on QAR data aiming at the problems, on one hand, the idea of big data is applied, and the more comprehensive multidimensional flight skill dimension is constructed by mining the correlation among indexes in the QAR data, so that the utilization rate of the data is improved; and on the other hand, a scientific flight skill scoring model is established, a visual multi-dimensional capability image is established for the pilot, and corresponding technical operation characteristic descriptions are matched with the pilot in a targeted manner according to the tendency level of each dimensional capability. The scheme has important practical significance for digging and evaluating the flight technical characteristics of the pilot and making and adjusting the pilot training plan in a targeted manner.
The embodiments in the present description are described in a progressive manner, each embodiment focuses on differences from other embodiments, and the same and similar parts among the embodiments are referred to each other. For the system disclosed by the embodiment, the description is relatively simple because the system corresponds to the method disclosed by the embodiment, and the relevant points can be referred to the method part for description.
The principles and embodiments of the present invention have been described herein using specific examples, which are provided only to help understand the method and the core concept of the present invention; meanwhile, for a person skilled in the art, according to the idea of the present invention, the specific embodiments and the application range may be changed. In view of the above, the present disclosure should not be construed as limiting the invention.
Claims (10)
1. A pilot flight skill representation method based on QAR data is characterized by comprising the following steps:
acquiring multiple flight task data of a set machine type in a set period;
determining effective data according to the multiple flight task data;
screening the effective data to construct a pilot flight skill portrait index system;
grading the flight skill dimension of the pilot by utilizing a grey correlation method according to the pilot flight skill portrait index system, and constructing a portrait model of the flight skill;
and constructing a radar chart of the multi-dimensional flight skill portrait of the pilot according to the portrait model.
2. The QAR data-based pilot flight skill representation method of claim 1, wherein the determining valid data from the multiple flight mission data specifically comprises:
preprocessing the flight task data to determine effective data; the preprocessing includes outlier processing and missing value processing.
3. The method for pilot flight skill representation based on QAR data as set forth in claim 1, wherein the screening of the valid data to construct a pilot flight skill representation index system specifically comprises:
screening the effective data according to the variance of the effective data to determine an effective evaluation data set; the valid evaluation data set comprises an original index;
standardizing the effective evaluation data set to obtain a standardized data set;
clustering the indexes in the standardized data set by using a principal component analysis method to determine new indexes;
screening the new indexes according to a linear transformation relation to determine flight skill dimensions; the linear transformation relation is a linear transformation relation between the original index and the new index;
and constructing a pilot flight skill image index system according to the effective evaluation data set and the flight skill dimension.
4. The QAR data-based pilot flight skill representation method of claim 3, wherein clustering the indicators in the normalized data set using principal component analysis to determine new indicators comprises:
calculating index correlation according to the standardized data set to obtain a correlation coefficient matrix;
determining the eigenvalue of the correlation coefficient matrix by using a Jacobi method;
determining a unit feature vector and an information contribution rate according to the feature value;
and clustering the indexes in the standardized data set according to the unit feature vector and the information contribution rate to determine a new index.
5. The QAR data-based pilot flight skill representation method as claimed in claim 4, wherein the step of scoring the flight skill dimension of the pilot by using a grey correlation method according to the pilot flight skill representation index system to construct the representation model of the flight skill specifically comprises the steps of:
determining the ideal value of the original index corresponding to the ideal value of the new index by utilizing the linear transformation relation according to the ideal value of the original index;
determining an absolute difference value according to the ideal value of the new index; the absolute difference is a difference between an ideal value of the new index and an actual value of the new index;
determining a correlation coefficient according to the absolute difference value; the correlation coefficient is a correlation coefficient between an ideal value of the new index and an actual value of the new index;
and determining the flight skill dimension score of the pilot according to the correlation coefficient to obtain an image model of the flight skill.
6. The QAR data-based pilot flight skill representation method of claim 5, wherein the building of the multidimensional flight skill representation radar chart of the pilot according to the representation model specifically comprises:
and constructing a multidimensional flight skill portrait radar chart of the pilot by utilizing a visual application program package according to the pilot flight skill dimension score in the portrait model.
7. A pilot flight skill representation system based on QAR data, comprising:
the acquisition module is used for acquiring multiple flight mission data of a set machine type in a set period;
the effective data determining module is used for determining effective data according to the multiple flight mission data;
the effective data screening module is used for screening the effective data and constructing a pilot flight skill portrait index system;
the scoring module is used for scoring the flight skill dimension of the pilot by utilizing a grey correlation degree method according to the pilot flight skill portrait index system and constructing a portrait model of the flight skill;
and the multidimensional flight skill portrait radar chart building module is used for building a multidimensional flight skill portrait radar chart of the pilot according to the portrait model.
8. The QAR data-based pilot flight skill representation system of claim 7, wherein the valid data filtering module specifically comprises:
the effective data screening unit is used for screening the effective data according to the variance of the effective data to determine an effective evaluation data set; the valid evaluation data set comprises an original index;
the standardization processing unit is used for carrying out standardization processing on the effective evaluation data set to obtain a standardized data set;
a new index determination unit for clustering the indexes in the normalized data set by using a principal component analysis method to determine new indexes;
the flight skill dimension determining unit is used for screening the new indexes according to a linear transformation relation and determining the flight skill dimension; the linear transformation relation is a linear transformation relation between the original index and the new index;
and the pilot flight skill portrait index system construction unit is used for constructing a pilot flight skill portrait index system according to the effective evaluation data set and the flight skill dimension.
9. The QAR data-based pilot flight skill representation system of claim 8, wherein the new indicator determination unit specifically comprises:
a correlation coefficient matrix determining subunit, configured to calculate index correlation according to the normalized data set, so as to obtain a correlation coefficient matrix;
an eigenvalue determination subunit, configured to determine an eigenvalue of the correlation coefficient matrix by using a jacobian method;
the unit characteristic vector and information contribution rate determining subunit is used for determining the unit characteristic vector and the information contribution rate according to the characteristic value;
and the new index determining subunit is used for clustering the indexes in the normalized data set according to the unit feature vector and the information contribution rate to determine a new index.
10. The QAR data-based pilot flight skill representation system of claim 9, wherein the scoring module specifically comprises:
the ideal value determining unit is used for determining the ideal value of the original index corresponding to the ideal value of the new index according to the ideal value of the original index by utilizing the linear transformation relation;
an absolute difference value determination unit for determining an absolute difference value from the ideal value of the new index; the absolute difference is a difference between an ideal value of the new index and an actual value of the new index;
a correlation coefficient determining unit, configured to determine a correlation coefficient according to the absolute difference; the correlation coefficient is a correlation coefficient between an ideal value of the new index and an actual value of the new index;
and the image model determining unit is used for determining the flight skill dimension score of the pilot according to the correlation coefficient to obtain an image model of the flight skill.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110721450.XA CN113312424A (en) | 2021-06-28 | 2021-06-28 | QAR data-based pilot flight skill portrait method and system |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN202110721450.XA CN113312424A (en) | 2021-06-28 | 2021-06-28 | QAR data-based pilot flight skill portrait method and system |
Publications (1)
| Publication Number | Publication Date |
|---|---|
| CN113312424A true CN113312424A (en) | 2021-08-27 |
Family
ID=77380652
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN202110721450.XA Pending CN113312424A (en) | 2021-06-28 | 2021-06-28 | QAR data-based pilot flight skill portrait method and system |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN113312424A (en) |
Cited By (2)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114611945A (en) * | 2022-03-15 | 2022-06-10 | 中国民用航空飞行学院 | Flight student competence portrait method based on multidimensional data |
| CN117972317A (en) * | 2024-03-29 | 2024-05-03 | 中国民用航空飞行学院 | Toughness portrait method oriented to pilot competence |
Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103792495A (en) * | 2014-01-29 | 2014-05-14 | 北京交通大学 | Method for evaluating battery performance based on Delphi method and grey relation theory |
| US20170210483A1 (en) * | 2016-01-21 | 2017-07-27 | Honeywell International Inc. | Evaluation of pilot performance using collected avionics system data |
| CN110334952A (en) * | 2019-07-05 | 2019-10-15 | 广西电力职业技术学院 | A kind of distribution network planning Post-assessment Method based on the improved grey model degree of association |
| CN111340388A (en) * | 2020-03-13 | 2020-06-26 | 中国民航大学 | Pilot operation quality evaluation method based on flight QAR data |
| CN111553411A (en) * | 2020-04-27 | 2020-08-18 | 中国民航大学 | Pilot risk portrait method based on multidimensional quantitative data |
-
2021
- 2021-06-28 CN CN202110721450.XA patent/CN113312424A/en active Pending
Patent Citations (5)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN103792495A (en) * | 2014-01-29 | 2014-05-14 | 北京交通大学 | Method for evaluating battery performance based on Delphi method and grey relation theory |
| US20170210483A1 (en) * | 2016-01-21 | 2017-07-27 | Honeywell International Inc. | Evaluation of pilot performance using collected avionics system data |
| CN110334952A (en) * | 2019-07-05 | 2019-10-15 | 广西电力职业技术学院 | A kind of distribution network planning Post-assessment Method based on the improved grey model degree of association |
| CN111340388A (en) * | 2020-03-13 | 2020-06-26 | 中国民航大学 | Pilot operation quality evaluation method based on flight QAR data |
| CN111553411A (en) * | 2020-04-27 | 2020-08-18 | 中国民航大学 | Pilot risk portrait method based on multidimensional quantitative data |
Cited By (3)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN114611945A (en) * | 2022-03-15 | 2022-06-10 | 中国民用航空飞行学院 | Flight student competence portrait method based on multidimensional data |
| CN114611945B (en) * | 2022-03-15 | 2024-02-02 | 中国民用航空飞行学院 | Flight student competence portrayal method based on multidimensional data |
| CN117972317A (en) * | 2024-03-29 | 2024-05-03 | 中国民用航空飞行学院 | Toughness portrait method oriented to pilot competence |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN110008442A (en) | A kind of pilot training's control method based on pilot's comprehensive ability evaluation | |
| CN113312424A (en) | QAR data-based pilot flight skill portrait method and system | |
| CN111553411A (en) | Pilot risk portrait method based on multidimensional quantitative data | |
| CN106503807B (en) | Improved RCM analysis method and mobile equipment integrity evaluation system based on same | |
| CN112396320A (en) | Tunnel collapse risk assessment method based on ISM and fuzzy Bayesian network | |
| CN113298431B (en) | Pilot competence portrayal method and system based on aviation QAR big data | |
| CN113344425B (en) | Flight quality monitoring method and system based on QAR data | |
| CN111340388A (en) | Pilot operation quality evaluation method based on flight QAR data | |
| CN113362604B (en) | Controller workload assessment method based on sector dynamic traffic characteristics | |
| CN116257681B (en) | Civil aviation event investigator recommendation method based on portrait technology | |
| CN114282792A (en) | Flight landing quality monitoring and evaluating method and system | |
| CN113793047A (en) | Pilot cooperative communication capacity evaluation method and device | |
| CN112800540B (en) | Aircraft engine load spectrum task segment modeling method based on Gaussian process regression | |
| Zhou et al. | The landing safety prediction model by integrating pattern recognition and Markov chain with flight data | |
| CN117235510A (en) | Joint roughness prediction method and training method of joint roughness prediction model | |
| CN104462787A (en) | RST/IAHP-based aeronautical maintenance human factor reliability assessment method | |
| CN109165854B (en) | Empty pipe operation efficiency grade evaluation method and device | |
| CN114398769B (en) | Automatic scoring acquisition method for unmanned helicopter flight control system | |
| Xiangzhang et al. | Risk Assessment and Prediction of Aircraft Fleet Hard Landing Based on QAR Exceedance Events | |
| CN115345408A (en) | Terminal area approach efficiency assessment method based on entropy weight method | |
| CN109190968B (en) | Empty pipe operation efficiency grade evaluation method and device | |
| CN114219302A (en) | Method for portraying wind competence of pilot facing to flight skill full life cycle | |
| CN108986554B (en) | Airspace sector crowding degree dynamic identification method based on fuzzy comprehensive judgment | |
| CN114117327A (en) | Abnormal flight recovery strategy evaluation method based on set-to-extension coupling | |
| CN108898890B (en) | Empty pipe operation efficiency grade evaluation method and device |
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 |