CN110390177A - The flying object that peels off determines method and device - Google Patents
The flying object that peels off determines method and device Download PDFInfo
- Publication number
- CN110390177A CN110390177A CN201910700625.1A CN201910700625A CN110390177A CN 110390177 A CN110390177 A CN 110390177A CN 201910700625 A CN201910700625 A CN 201910700625A CN 110390177 A CN110390177 A CN 110390177A
- Authority
- CN
- China
- Prior art keywords
- flying object
- peels
- test point
- mean value
- flying
- 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.)
- Granted
Links
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/20—Design optimisation, verification or simulation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Computer Hardware Design (AREA)
- Evolutionary Computation (AREA)
- Geometry (AREA)
- General Engineering & Computer Science (AREA)
- General Physics & Mathematics (AREA)
- Traffic Control Systems (AREA)
Abstract
This disclosure relates to which one kind peels off, flying object determines method and device, which comprises obtains operating parameter of at least one flying object when reaching default test point in the predetermined landing stage;Determine that model determines the flying object that peels off at least one described flying object using the operating parameter and the flying object that peels off corresponding with predetermined landing stage.By above method, the disclosure can quickly and accurately determine the flying object that peels off at least one flying object, also, the disclosure is to carry out comprehensive descision to the flying object that peels off based on whole industry QAR data and multiple parameters, avoids the dependence to subjective experience.
Description
Technical field
This disclosure relates to which technical field of data processing more particularly to one kind peel off, flying object determines method and device.
Background technique
It is shown according to Boeing statistical report in 2017, takeoff phase only accounts for the 2% of entire mission phase, and landing phases are only
The 4% of entire mission phase is accounted for, but during 2008 to 2017, the ratio that disastrous accident occurs for takeoff phase accounts for all fatal
The 14% of accident, the ratio that disastrous accident occurs for landing phases account for the 49% of all disastrous accidents, it can be seen that takeoff and landing
It is the entire flight course risk highest stage.Recently as development in science and technology, Aviation accident rates decline year by year, by 2019
Year, China's transport aviation had been carried out lasting safe flight 102 months, and the record is also constantly increasing by the end of February.Although not yet
There is disastrous accident, but may cause the unsafe accident of disastrous accident frequently, according to record, January nineteen ninety in April, 2008
Between, wiping tail event 24 occurred altogether and rises civil aviaton, China flight takeoff phase.Safety is the theme of civil aviaton always, how further to be mentioned
High safety is horizontal, and the safety accident for especially taking precautions against the landing phases that take off in advance is the major issue for needing constantly to think deeply.
In order to promote aviation safety level, China Civil Aviation office was provided from January 1st, 1998, was registered simultaneously within Chinese territory
The carrier of operation should install quick access recorder (quick access recorder, QAR) or equivalent of the apparatus, and
QAR data are applied directly to civil aviaton's flight quality monitoring (Flight Operations Quality since 2000
Assurance, FOQA), to promote flight safety management level.2017, China Civil Aviation flight quality monitoring base station was formally thrown
Enter operation, daily receiving country Nei Ge airline is more than the QAR data of 3100 airplanes, more than 16000 a flights, to monitor boat
Class's flight condition, finds operation risk in time.QAR data cover airplane flight data very rich, including time, speed
Degree, posture, position, aircraft engine, APU auxiliary power unit, flight control system, fuel system, blank pipe information, rises and falls at height
Frame, inertial navigation system, gear-box etc..QAR data record parameters of the aircraft in flight course, can continuous whole ground it is anti-
The virtual condition and various signs of aircraft system in operation are reflected, is the weight that data science is applied in safety of civil aviation and operation field
Want basis.
The appearance of QAR data provides new application direction for aviation safety, but current application is concentrated mainly on and transfinites
Affair alarm, for example, the relevant technologies are usually to preset the risk of certain unitary variants according to aircraft supplier or experience
Then threshold value carries out each variable compared with threshold value one by one, more than the flight for judging the flight after preset threshold value
There are security risks.The relevant technologies excessively rely on subjective experience, and man's activity is larger, and given threshold value is usually compared
Loosely, reduce the pre-alerting ability to safety in some sense;Secondly, this method is mainly the comparison of unitary variant, lack
Between variable the considerations of relevance;In addition, existing application is mostly based on the data of single airline or single airport, do not have
There is the application level for covering entire industry, adaptibility to response is lacked to whole deviation.
Summary of the invention
In view of this, the present disclosure proposes one kind flying objects that peels off to determine method, which comprises obtain at least one
Operating parameter of the flying object when reaching default test point in the predetermined landing stage;Make a reservation for using the operating parameter and with described
The corresponding flying object that peels off of depression of order section determines that model determines the flying object that peels off at least one described flying object.
It is described using the operating parameter and corresponding with the predetermined landing stage in a kind of possible embodiment
The flying object that peels off determines that model determines the flying object that peels off at least one described flying object, comprising: determines the predetermined landing
Estimation in stage under N number of default test point peels off flying object;By be confirmed as estimation peel off flying object number not less than M
Flying object is determined as the flying object that peels off, wherein M, N are natural number, M≤N.
In a kind of possible embodiment, the method also includes: it obtains multiple flying objects and is arrived in the predetermined landing stage
Up to the historical operating parameter of default test point;Data are carried out to the historical operating parameter to analyze to obtain data characteristics, the number
According to the state and correlation that feature includes between the density function of each historical operating parameter, each historical operating parameter;According to
The data characteristics establishes the flying object that peels off and determines model.
It is described that historical operating parameter progress data are analyzed to obtain data spy in a kind of possible embodiment
Sign, comprising: establish the scatterplot matrices between two two parameters using the historical operating parameter;It is true according to the scatterplot matrices
The fixed data characteristics.
In a kind of possible embodiment, the flying object that peels off is established according to the data characteristics and determines model, comprising: In
There are historical operating parameters to be rendered as discrete state, or, there are the feelings that multimodal characteristic is presented in the density function of historical operating parameter
Under condition, the flying object that peels off described in determination determines that model is multiclass Clustering Model;Or convergence is presented in all historical operating parameters
In the case that single-peak response is presented in the density function of state and all historical operating parameters, the flying object that peels off described in determination is determined
Model is single class Clustering Model.
In a kind of possible embodiment, there are in the case where correlation between multiple historical operating parameters, utilize
Other historical operating parameters of one of multiple historical operating parameters with correlation and not correlation are as training
Data.
It is described that model is determined according to the data characteristics foundation flying object that peels off in a kind of possible embodiment, also
Include: under the premise of avoiding over-fitting and poor fitting, using the training data to the flying object that peels off determine model into
Row training;The model parameter of model, which is adjusted, to be determined to the flying object that peels off, when the flying object that peels off determined in training
Number when accounting for all flying object numbers and reaching the first ratio, determine described in the flying object that peels off determine the final mask ginseng of model
Number, and the flying object that peels off after being trained according to the final mask parameter determines model.
In a kind of possible embodiment, first ratio is between 5% to 10%.
It is described default in the case where the predetermined landing stage is takeoff phase in a kind of possible embodiment
Test point includes the first test point, and first test point is the position of the moment of taking off of flying object, the operating parameter
Including distance of taking off, flying speed, the liftoff elevation angle, liftoff elevation angle change rate mean value, liftoff elevation angle change rate standard deviation, it is liftoff hang down
Any one or more of straight speed, flying speed change rate mean value, flying speed change rate standard deviation, wherein described to take off
Distance indicates that flying object flies to horizontal distance when flying object reaches default safe altitude from the outset, and the flying speed indicates
In the horizontal velocity of flying object main wheel liftoff instant flying object, the liftoff elevation angle indicates to fly in flying object main wheel liftoff instant
The elevation angle of object, the liftoff elevation angle change rate mean value indicate that each second elevation angle changes equal in flying object main wheel liftoff front and back n seconds
Value, the liftoff elevation angle change rate standard deviation indicate the standard deviation of each second liftoff front and back n seconds elevation angle of flying object main wheel variation, institute
State vertical speed of the liftoff vertical speed expression in flying object main wheel liftoff instant flying object, the flying speed change rate mean value
Indicate the mean value of each second liftoff front and back n seconds vertical speed of flying object main wheel variation, the flying speed change rate standard deviation table
Show the standard deviation of each second liftoff front and back n seconds vertical speed of flying object main wheel variation, wherein n > 0.
It is described default in the case where the predetermined landing stage is into the nearly stage in a kind of possible embodiment
Test point includes the second test point, third test point and the 4th test point, and second test point is that flying object reaches instrument supply gas
Position where when as the stabilization under condition IMC into nearly detection height, second test point is that flying object reaches visual meteorological
Position where when stabilization under condition VMC is into nearly detection height, when the third test point is that flying object is in five side of airport
Stabilization into nearly detection height when where position.
In a kind of possible embodiment, the height of second test point is 1000 feet, the third test point
Height be 500 feet, the height of the 4th test point is 300 feet.
In a kind of possible embodiment, in the case where the predetermined landing stage is into the nearly stage, the operation
Parameter includes pitch angle mean value, course mean value, relative velocity mean value, vertical speed mean value, each engine low pressure rotor standard
Any one or more of poor, each engine high pressure rotor standard deviation, wherein the pitch angle mean value indicates flying object height
The average value of each second m seconds pitch angle in default test point front and back is reached, it is pre- that the course mean value indicates that flying object height reaches
If the average value of each second m seconds course variation before and after test point,
The relative velocity mean value indicates that flying object height reaches each second m seconds air speed in default test point front and back and subtracts reference
The average value of speed, the vertical speed mean value indicate that flying object height reaches the default each second m seconds inertia in test point front and back and hangs down
The average value of straight speed, the engine low pressure rotor standard deviation indicate flying object m seconds before and after the default test point of height arrival
The standard deviation of the low pressure rotor velocity variations of each second engine, the engine high pressure rotor standard deviation indicate to reach when height
The standard deviation of the low pressure rotor velocity variations of each second m seconds engine before and after default test point, wherein m > 0.
In a kind of possible embodiment, the predetermined landing stage be landing it is flat float the stage in the case where, it is described
Default test point includes the 5th test point, the 6th test point, the 7th test point, wherein the 5th test point be flying object from
Position where when pitching rule is revised as evening up rule, the 6th test point are as defined in flying object reaches under stable condition
Position where when flare out altitude, the 7th test point, which receives to propaganda directed to communicate automatically for flying object, reminds pilot to withdraw thrust hand
Position where when handle.
In a kind of possible embodiment, the height of the 5th test point is 50 feet, the 6th test point
Height is 30 feet, and the height of the 7th test point is 20 feet.
In a kind of possible embodiment, the predetermined landing stage be landing it is flat float the stage in the case where, it is described
Operating parameter includes Inertial Vertica1 Speed mean value, pitch angle mean value, air speed mean value, vertically overloads mean value, ground distance and ground connection
Time any one and it is a variety of, wherein before the Inertial Vertica1 Speed mean value indicates that flying object height reaches default test point
The average value of each second k seconds Inertial Vertica1 Speed afterwards, before the pitch angle mean value indicates that flying object height reaches default test point
The average value of each second k seconds pitch angle afterwards, the air speed mean value indicate that flying object height reaches default test point front and back k seconds often
The average value of one second air speed (IAS), the vertical overload mean value indicate that flying object height reaches default test point front and back k seconds often
The average value of one second vertical overload VRTG presets the horizontal distance of grounding point to pick-up point, ground connection described in the ground distance
Time indicates time of the flying object from the default grounding point to pick-up point, wherein k > 0.
According to another aspect of the present disclosure, it proposes one kind to peel off flying object determining device, described device includes: first to obtain
Modulus block, for obtaining operating parameter of at least one flying object when reaching default test point in the predetermined landing stage;Determine mould
Block is connected to the acquisition module, for utilizing the operating parameter and the flight that peels off corresponding with the predetermined landing stage
Object determines that model determines the flying object that peels off at least one described flying object.
According to another aspect of the present disclosure, it provides one kind to peel off flying object determining device, comprising: processor;For depositing
Store up the memory of processor-executable instruction;Wherein, the processor is configured to executing the above method.
According to another aspect of the present disclosure, a kind of non-volatile computer readable storage medium storing program for executing is provided, is stored thereon with
Computer program instructions, wherein the computer program instructions realize the above method when being executed by processor.
By above method and device, at least one available flying object of the disclosure reaches default in the predetermined landing stage
Operating parameter when test point determines mould using the operating parameter and the flying object that peels off corresponding with predetermined landing stage
Type determines the flying object that peels off at least one described flying object.The disclosure can quickly and accurately determine at least one flight
The flying object that peels off in object, also, the disclosure is to be integrated based on whole industry QAR data and multiple parameters to the flying object that peels off
Judgement, avoids the dependence to subjective experience.
According to below with reference to the accompanying drawings to detailed description of illustrative embodiments, the other feature and aspect of the disclosure will become
It is clear.
Detailed description of the invention
Comprising in the description and constituting the attached drawing of part of specification and specification together illustrates the disclosure
Exemplary embodiment, feature and aspect, and for explaining the principles of this disclosure.
Fig. 1 shows the flow chart that method is determined according to the flying object that peels off of one embodiment of the disclosure.
Fig. 2 shows the flow charts that method is determined according to the flying object that peels off of one embodiment of the disclosure.
Fig. 3 a shows the schematic diagram in the historical operating parameter of the same model flying object of takeoff phase acquisition.
Fig. 3 b shows the schematic diagram in takeoff phase using the scatterplot matrices of historical operating parameter foundation.
Fig. 3 c, which is shown, determines schematic diagram that the parameter of model is adjusted to the flying object that peels off of takeoff phase.
Fig. 3 d, which is shown, determines that model carries out the determining schematic diagram of flying object that peels off using the flying object that peels off of mission phase.
Fig. 4 a shows the schematic diagram in the historical operating parameter of the same model flying object obtained into the nearly stage.
Fig. 4 b shows the schematic diagram in the scatterplot matrices established into the nearly stage using historical operating parameter.
Fig. 4 c, which is shown, determines schematic diagram that the parameter of model is adjusted to the flying object that peels off into the nearly stage.
Fig. 4 d, which is shown, determines that model carries out the flying object that peels off in the second test point using the flying object that peels off into the nearly stage
Determining schematic diagram.
Fig. 4 e, which is shown, determines that model carries out the flying object that peels off in third test point using the flying object that peels off into the nearly stage
Determining schematic diagram.
Fig. 4 f, which is shown, determines that model carries out the flying object that peels off in the 4th test point using the flying object that peels off into the nearly stage
Determining schematic diagram.
Fig. 4 g shows the schematic diagram that the flying object that peels off is determined in conjunction with multiple predetermined detection points into the nearly stage.
Fig. 5 a shows the schematic diagram in the historical operating parameter of the same model flying object of the flat stage acquisition of floaing of landing.
Figure 5b shows that utilize the schematic diagram of the scatterplot matrices of historical operating parameter foundation in the landing flat stage of floaing.
Fig. 5 c shows the flying object that peels off for floaing the stage flat to landing and determines the schematic diagram that the parameter of model is adjusted.
Fig. 5 d shows that determine that model in the 5th test point peel off using the flying object that peels off in landing flat stage of floaing winged
The schematic diagram that row object determines.
Fig. 5 e shows that determine that model in the 6th test point peel off using the flying object that peels off in landing flat stage of floaing winged
The schematic diagram that row object determines.
Fig. 5 f shows that determine that model in the 7th test point peel off using the flying object that peels off in landing flat stage of floaing winged
The schematic diagram that row object determines.
Fig. 5 g shows that determine that model in the 8th test point peel off using the flying object that peels off in landing flat stage of floaing winged
The schematic diagram that row object determines.
Fig. 5 h shows the schematic diagram that the flying object that peels off is determined in conjunction with multiple predetermined detection points in landing flat stage of floaing.
Fig. 6 shows the block diagram of the flying object determining device that peels off according to one embodiment of the disclosure.
Fig. 7 shows the block diagram of the flying object determining device that peels off according to one embodiment of the disclosure.
Specific embodiment
Various exemplary embodiments, feature and the aspect of the disclosure are described in detail below with reference to attached drawing.It is identical in attached drawing
Appended drawing reference indicate element functionally identical or similar.Although the various aspects of embodiment are shown in the attached drawings, remove
It non-specifically points out, it is not necessary to attached drawing drawn to scale.
Dedicated word " exemplary " means " being used as example, embodiment or illustrative " herein.Here as " exemplary "
Illustrated any embodiment should not necessarily be construed as preferred or advantageous over other embodiments.In addition, in order to better illustrate the disclosure, In
Numerous details is given in following description.It will be appreciated by those skilled in the art that without certain specific
Details, the disclosure equally can be implemented.In some instances, for method well known to those skilled in the art, means, element and
Circuit is not described in detail, in order to highlight the purport of the disclosure.
Referring to Fig. 1, Fig. 1 shows the flow chart for determining method according to the flying object that peels off of one embodiment of the disclosure.
The method can be applied in terminal and/or server.
As shown in Figure 1, which comprises
Step S110 obtains operating parameter of at least one flying object when reaching default test point in the predetermined landing stage;
Step S120 determines model using the operating parameter and the flying object that peels off corresponding with predetermined landing stage
Determine the flying object that peels off at least one described flying object.
By above method, at least one available flying object of the disclosure reaches default test point in the predetermined landing stage
When operating parameter, determine that model determines using the operating parameter and the flying object that peels off corresponding with predetermined landing stage
The flying object that peels off at least one described flying object.The disclosure can quickly and accurately determine at least one flying object
Peel off flying object, also, the disclosure is to carry out comprehensive descision to the flying object that peels off based on whole industry QAR data and multiple parameters,
Avoid the dependence to subjective experience.
Determining that method has determined the flight that peels off at least one flying object using the flying object that peels off described in the disclosure
In the case where object, indicate that takeoff and landing performance of the flying object near default test point and group preset rising near test point at this
It is larger to drop aberrations in property, needs to be adjusted.After obtaining result, control centre can be sent result to and referred to,
Or can give a warning, to remind staff to pay attention to averting risks.
In a kind of possible embodiment, the disclosure can obtain the operating parameter of one or more flying objects in real time,
To whether be that the flying object that peels off judges to the one or more flying object.Also one or more of available a period of time
The operating parameter of a flying object, to be judged.In brief, the flying object that peels off that the disclosure proposes determines that method can answer
Used in hard real time scene and weak projectivity scene.
In a kind of possible embodiment, the predetermined landing stage may include takeoff phase, into the nearly stage and fall
Horizon floats the stage.
In a kind of possible embodiment, default test point can be with reference to right in airline flight machine set technology handbook
It is determined in the technical description for landing of taking off.
For the different predetermined landing stages, different default test points can be determined, the number of default test point can be
One, be also possible to it is multiple, specific number can according to analysis demand determine, the disclosure to it is specific preset test point not do
It limits, also, also without limitation to the number of default test point.
In a kind of possible embodiment, the measurement using height as default test point can choose.
It highly measures, avoids using radar altitude, because radar is high for example, pressure altitude can be selected to be used as into the nearly stage
Degree is larger by landform and effect on building fluctuation;Landing the flat stage of floaing can select radar altitude to measure as height, because into
Radar altitude is influenced smooth fluctuations by barrier after, can be used as elevation references.
In one example, the default test point of takeoff phase can choose in ground level.
In one example, into the default test point in nearly stage may include the stabilization under instrument meteoro logical condition, IMC (IMC) into
Stabilization when stabilization under nearly detection height, visual meteorological condition (VMC) (VMC) is in five side of airport into nearly detection height, aircraft is into close
Detection height etc..
In one example, when the default test point in landing flat stage of floaing may include pitching rule and be revised as evening up rule
Height, defined flare out altitude, the height propagandaed directed to communicate when pilot being reminded to withdraw thrust handle automatically etc. under stable condition.
In a kind of possible embodiment, the operating parameter in different landing stages can be determines according to actual conditions.
In one example, operating parameter can be divided into two class of direct measurement parameter and calculating parameter, and direct measurement parameter is airborne QAR
The direct measurement data that data provide, such as speed, pitch angle, climbing;Calculating parameter can be for based on direct measurement parameter
The calculated value calculated, such as average value, standard deviation etc..
A recapitulative introduction has been done to default test point, operating parameter etc. above, will be described in more detail later.
It should be noted that those skilled in the art can according to actual needs to default test point, operating parameter etc. into
Row selection.
In a kind of possible embodiment, the flying object can be various types of aircrafts.Such as it currently transports
The flight of battalion, including a plurality of types of aircrafts.Certainly, in other implementations, it is also possible to other, for example, it is also possible to
It is business or the unmanned plane that non-commercial uses.
In a kind of possible embodiment, step S120 using the operating parameter and with the predetermined landing stage pair
The flying object that peels off answered determines the flying object that peels off at least one described flying object of model determination, may include: described in determination
Estimation in the predetermined landing stage under N number of default test point peels off flying object;It will be confirmed as estimation to peel off the number of flying object
Flying object not less than M is determined as the flying object that peels off, wherein M, N are natural number, M≤N.
The disclosure can set the minimum detection number (for example, M) for the flying object that persistently peels off, according to actual use demand
In all default test points in predetermined landing stage, if the number that peels off that certain flying objects are detected is not less than minimum detection
Number indicates that these flying objects are controlled not over flight by flying object then by these flying objects labeled as the flying object that persistently peels off
Takeoff and landing performance adjustment and recovery, the flying object of these flying objects that are confirmed as peeling off for opposite normal flight object group, continues
The risk that problem occurs for the flying object that peels off is higher, needs to pay close attention to.
Referring to Fig. 2, Fig. 2 shows the flow charts for determining method according to the flying object that peels off of one embodiment of the disclosure.
In a kind of possible embodiment, the method can also include:
Step S130 obtains the historical operating parameter that multiple flying objects reach default test point in the predetermined landing stage;
Step S140 carries out data to the historical operating parameter and analyzes to obtain data characteristics, and the data characteristics includes
State and correlation between the density function of each historical operating parameter, each historical operating parameter;
Step S150 establishes the flying object that peels off according to the data characteristics and determines model.
By above method, the available multiple flying objects of the disclosure reach going through for default test point in the predetermined landing stage
History operating parameter carries out data to the historical operating parameter and analyzes to obtain data characteristics, and the data characteristics includes each goes through
State and correlation between the density function of history operating parameter, each historical operating parameter are established according to the data characteristics
The flying object that peels off determines model.The flying object that peels off that method according to the present disclosure determines determines model, can be quick and precisely
Ground obtains the flying object that peels off being detected in flying object.
In a kind of possible embodiment, the historical operating parameter can for one month, one week or one day or other
The history QAR data of any time period.
In a kind of possible embodiment, step S140 carries out data to the historical operating parameter and analyzes to obtain data
Feature may include: the scatterplot matrices established between two two parameters using the historical operating parameter;According to the scatter plot
Matrix determines the data characteristics.
In a kind of possible embodiment, step S150 establishes the flying object that peels off according to the data characteristics and determines mould
Type may include:
There are historical operating parameters to be rendered as discrete state, or, there are the presentation of the density function of historical operating parameter is more
In the case where peak character, the flying object that peels off described in determination determines that model is multiclass Clustering Model;Or in all historical operating parameters
In the case that single-peak response is presented in the density function of presentation convergence state and all historical operating parameters, peel off described in determination
Flying object determines model for single class Clustering Model.
In a kind of possible embodiment, single class Clustering Model for example may include One-class SVM, robust association side
Difference etc..In a kind of possible embodiment, multiclass Clustering Model for example may include DBSCAN, GMM etc..Certainly, it singly birdss of the same feather flock together
More than that, above is the illustrative description and citing carried out to it, art technology for class model and multiclass Clustering Model
Personnel can choose other single class Clustering Models and multiclass Clustering Model, in this regard, the disclosure is with no restrictions.
By above method, the disclosure can be the case where the data characteristics of multiple historical operating parameters meets different condition
Under, determine that different types of Clustering Model is flown as peeling off without determining model, so that the adaptability of varying environment is improved, so that
The flying object that peels off described in the disclosure determines that method can be applied to a variety of different situations.
It, can be with there are in the case where correlation between multiple historical operating parameters in a kind of possible embodiment
Using one of multiple historical operating parameters with correlation and not other historical operating parameters of correlation as
Training data.In a kind of possible embodiment, correlation can refer to that the Pearson correlation coefficient between parameter is not less than
0.8。
By determining the correlation between parameter, and utilize one of multiple historical operating parameters with correlation
And model number of dimensions does not can be effectively reduced as training data in other historical operating parameters of correlation, to save operation
Resource improves arithmetic speed.
In a kind of possible embodiment, step S150 establishes the flying object that peels off according to the data characteristics and determines mould
Type can also include:
Under the premise of avoiding over-fitting and poor fitting, model is determined to the flying object that peels off using the training data
It is trained;The model parameter of model, which is adjusted, to be determined to the flying object that peels off, when the flight that peels off determined in training
When the number of object accounts for all flying object numbers and reaches the first ratio, determine described in the flying object that peels off determine the final mask ginseng of model
Number, and the flying object that peels off after being trained according to the final mask parameter determines model.
By being determined using the training data to the flying object that peels off under the premise of avoiding over-fitting and poor fitting
Model is trained, and the available accurately flying object that peels off of the disclosure determines model, thus, in sentencing for the flying object that peel off
Periodically, available accurate result.
In a kind of possible embodiment, first ratio is between 5% to 10%.By by the first ratio setting
Between 5% to 10%, the demand of engineer application can satisfy, so that more accurate to the judgement for the flying object that peels off.
The different phase in predetermined landing stage will be introduced respectively below.
In takeoff phase:
It is described default in the case where the predetermined landing stage is takeoff phase in a kind of possible embodiment
Test point may include the first test point, and first test point is the position of the moment of taking off of flying object.In flying object
Takeoff phase, flying object and ground face contact, height off the ground be 0 foot.
In a kind of possible embodiment, the operating parameter may include take off distance (DIST_TAKEOFF), from
Ground speed (IAS_LIFTOFF), the liftoff elevation angle (PITCH_LIFTOFF), liftoff elevation angle change rate mean value (PITCH_RATE_
AVE), liftoff elevation angle change rate standard deviation (PITCH_RATE_STD), liftoff vertical speed (IVV_LIFTOFF), flying speed
Any one or more of change rate mean value (IVV_RATE_AVE), flying speed change rate standard deviation (IVV_RATE_STD),
Wherein,
The distance of taking off indicates that flying object flies to horizontal distance when flying object reaches default safe altitude from the outset,
The flying speed indicates the horizontal velocity in flying object main wheel liftoff instant flying object, and the liftoff elevation angle is indicated in flying object
The elevation angle of main wheel liftoff instant flying object, the liftoff elevation angle change rate mean value indicate every in flying object main wheel liftoff front and back n seconds
The mean value of one second elevation angle variation, the liftoff elevation angle change rate standard deviation indicate that flying object main wheel liftoff front and back n seconds each seconds face upward
The standard deviation of angle variation, the liftoff vertical speed indicates the vertical speed in flying object main wheel liftoff instant flying object, described
Flying speed change rate mean value indicates the mean value of each second liftoff front and back n seconds vertical speed of flying object main wheel variation, described liftoff
Percentage speed variation standard deviation indicates the standard deviation of each second liftoff front and back n seconds vertical speed of flying object main wheel variation, wherein n >
0。
In a kind of possible embodiment, the specific size of n can determines according to actual conditions, in one example, n
It can be 5s.
The example in the operating parameter of takeoff phase is presented above, it should be appreciated that, those skilled in the art can be with
Increase according to the actual situation, reduce operating parameter, in this regard, the disclosure is without limitation.
The determination of model training and the flying object that peels off in takeoff phase is illustrated below.
Please refer to Fig. 3 a, Fig. 3 a shows showing for the historical operating parameter of the same model flying object obtained in takeoff phase
It is intended to.
According to step S130, the disclosure can obtain the same model flying object as shown in Figure 3a obtained in takeoff phase
Historical operating parameter.
It is to be understood that this is introduced for sentencing the flying object of same model, still, the disclosure more than that,
Determination for the flying object that peels off of different types of flying object, the disclosure are equally applicable.
As shown in Figure 3a, in one example, the disclosure has selected 28 flying objects (number No-X0001-X0028)
Historical operating parameter, they include: take off distance (DIST_TAKEOFF), flying speed (IAS_LIFTOFF), the liftoff elevation angle
(PITCH_LIFTOFF), liftoff elevation angle change rate mean value (PITCH_RATE_AVE), liftoff elevation angle change rate standard deviation
(PITCH_RATE_STD), liftoff vertical speed (IVV_LIFTOFF), flying speed change rate mean value (IVV_RATE_AVE),
Flying speed change rate standard deviation (IVV_RATE_STD).
Here and hereinafter, the disclosure does not introduce the unit of historical operating parameter, operating parameter, it is clear that
It is that the unit of each parameter can be determines according to actual conditions.
After obtaining historical operating parameter, according to step S140, the disclosure can be counted the historical operating parameter
Data characteristics is obtained according to analysis, the data characteristics includes the density function of each historical operating parameter, each history run ginseng
State and correlation between number.
For example, can use the scatterplot matrices that the historical operating parameter is established between two two parameters;It is dissipated according to described
Point diagram matrix determines the data characteristics.
Fig. 3 b is please referred to, Fig. 3 b shows the signal in takeoff phase using the scatterplot matrices of historical operating parameter foundation
Figure.As shown in Figure 3b, on the diagonal line of scatterplot matrices, the density function of each parameter has single-peak response and (only has
One wave crest), it is similar to normal distribution;Also, there is apparent data to converge characteristic between parameter and parameter, and without discrete
Parameter.
According to step S150, the flying object that peels off can be established according to the data characteristics and determine model.
In one example, discrete parameter is not present in historical operating parameter, then the flying object test problems that peel off can be seen
At being single class clustering problem, i.e., normal flight species and peels off flight species, can choose corresponding single class clustering algorithm and carry out
Peel off flight analyte detection, for example, One-class SVM can be selected.
Please refer to Fig. 3 c, Fig. 3 c, which is shown, determines that the parameter of model was adjusted shows to the flying object that peels off of takeoff phase
It is intended to.
Herein, for convenience, selection is taken off distance DIST_TAKEOFF and flying speed IAS_LIFTOFF two
A parameter is analyzed, it should explanation, multi-parameter (including except the distance DIST_TAKEOFF and flying speed IAS_ that takes off
Other parameters other than LIFTOFF) when the model and training method it is equally applicable.
Based on the data characteristics of takeoff phase each historical operating parameter, the disclosure illustratively selects One-Class
SVM algorithm carries out clustering learning, wherein kernel function is illustratively selected radial basis function (RBF), and to two parameters of algorithm
Nu (for providing the training error coboundary of SVM algorithm) and gamma (for adjusting the scale parameter of radial basis function) is carried out
Parameter adjustment.Under the premise of avoiding over-fitting and poor fitting, parameter nu and parameter gamma are adjusted, until being adjusted to
The quantity accounting of flying object of peeling off is best between 5%~10%.
As shown in Figure 3c, black sideline (ellipse etc.) is the classification boundaries that algorithm learns, and is instruction outside classification boundaries
Practice the flying object that peels off being marked in the process, is the normal flight object being marked in training process in classification boundaries.It can be with
See, parameter nu is bigger, and classification boundaries are got over inconocenter and collected;Parameter gamma can be used to adjust the shape of classification boundaries, the parameter
Bigger, classification boundaries are more bent, but the parameter is excessive, over-fitting easily occur (as shown in the lower right corner in Fig. 3 c).Over-fitting and
Poor fitting can all influence abnormality detection as a result, so the disclosure parameter adjustment in avoided.It can be seen that and work as from Fig. 3 c
When parameter nu is 0.1 and parameter gamma is 5, the flying object that peels off of takeoff phase determines that model has preferable performance.
After determining that model training is good to the flying object that peels off, i.e., using peel off flying object determine model to need carry out from
The flying object of group's judgement is judged.
Please refer to Fig. 3 d, Fig. 3 d, which is shown, determines that the model flying object that peel off is true using the flying object that peels off of mission phase
Fixed schematic diagram.
As shown in Figure 3d, the flying object that peels off is determined that the parameter nu of model is set as 0.1, sets 5 for parameter gamma,
By operating parameter (distance of taking off (DIST_TAKEOFF), flying speed (IAS_ of at least one flying object (such as 480)
LIFTOFF), the liftoff elevation angle (PITCH_LIFTOFF), liftoff elevation angle change rate mean value (PITCH_RATE_AVE), the liftoff elevation angle
Change rate standard deviation (PITCH_RATE_STD), liftoff vertical speed (IVV_LIFTOFF), flying speed change rate mean value
(IVV_RATE_AVE), one or more of flying speed change rate standard deviation (IVV_RATE_STD)) input the flight that peels off
Object determines in model, and the flying object that can determine to peel off is about 43 (the dark points in Fig. 3 d), Zhan Suoyou training data
9.03% or so.
In the above examples, for takeoff phase, the default test point of disclosure selection is only 1, therefore, for upper
Example, 43 flying objects that peel off determined are final result.
It should be noted that in example in this example and later, (visualization can only be used due to visualizing limitation
3 dimensions), other parameters (removing DIST_TAKEOFF, IAS_LIFTOFF) can not embody in the figure still, flying object in the diagram
It whether is that the flying object that peels off can be embodied in figure.For example, each flying object can be marked by 8 parameter input models
It is denoted as normal or peels off, embody regular flight (light color) in three-dimensional figure;Peel off flight (dark color).
For into the nearly stage:
According to the technical manual of airline crew, when meeting following all criterion, it is considered as stablizing into nearly rank
Section:
1) aircraft is in correctly flight boat diameter;2) it only needs slightly to change pitching and boat for the boat diameter that keeps correctly flying
To;3) aircraft is in approach speed;4) aircraft is in correct landing configuration;5) deflection ratio is not more than some setting value;6) it pushes away
Power is set up to be adapted with the form of aircraft;7) all letters enable and checklist has executed.
It is described default in the case where the predetermined landing stage is into the nearly stage in a kind of possible embodiment
Test point may include the second test point, third test point and the 4th test point, and second test point can reach for flying object
Position where when stabilization under to instrument meteoro logical condition, IMC IMC is into nearly detection height, second test point can be flying object
Position where when reaching stabilization under visual meteorological condition (VMC) VMC into nearly detection height, the third test point can be flight
Position where when stabilization when object is in five side of airport is into nearly detection height.
In a kind of possible embodiment, the height of second test point can be 1000 feet, the third inspection
The height of measuring point can be 500 feet, and the height of the 4th test point can be 300 feet.
In a kind of possible embodiment, in the case where the predetermined landing stage is into the nearly stage, the operation
Parameter may include pitch angle mean value (PITCH_ave), course mean value (Head_diff), relative velocity mean value
(IASminusVREF), vertical speed mean value (IVV_ave), each engine low pressure rotor standard deviation, each engine high pressure
Any one or more of rotor standard deviation, wherein before the pitch angle mean value indicates that flying object height reaches default test point
The average value of each second m seconds pitch angle afterwards, the course mean value indicate that flying object height reaches default test point front and back m seconds often
The average value of one second course variation, the relative velocity mean value indicate before and after flying object height reaches default test point m seconds it is each
Second air speed subtracts the average value of reference velocity, and the vertical speed mean value indicates that flying object height reaches m before and after default test point
The average value of each second Inertial Vertica1 Speed of second, the engine low pressure rotor standard deviation indicate that flying object is preset when height reaches
The standard deviation of the low pressure rotor velocity variations of each second m seconds engine, the engine high pressure rotor standard deviation before and after test point
Indicate the standard deviation of the low pressure rotor velocity variations of each second m seconds engine before and after height reaches default test point, wherein m
> 0.
In a kind of possible embodiment, the specific size of m can determines according to actual conditions, in one example, m
It can be 5s.
The example in the operating parameter into the nearly stage is presented above, it should be appreciated that, those skilled in the art can be with
Increase according to the actual situation, reduce operating parameter, in this regard, the disclosure is without limitation.
The determination in the model training into the nearly stage and the flying object that peels off is illustrated below.
Please refer to Fig. 4 a, Fig. 4 a shows the historical operating parameter in the same model flying object obtained into the nearly stage and shows
It is intended to.
According to step S130, the disclosure can obtain as shown in fig. 4 a in the same model flying object obtained into the nearly stage
Historical operating parameter.
As shown in fig. 4 a, in one example, the disclosure has selected 27 flying objects (number is No-X0001~X0027)
Historical operating parameter.
In a kind of possible embodiment, in the case where the predetermined landing stage is into the nearly stage, the operation
Parameter may include pitch angle mean value (PITCH_ave), course mean value (Head_diff), relative velocity mean value
(IASminusVREF), vertical speed mean value (IVV_ave), each engine low pressure rotor standard deviation, each engine high pressure
Rotor standard deviation.Wherein, in this example, as shown in fig. 4 a, the disclosure is obtained starts for m seconds before and after height reaches test point
The standard deviation (N11_std) of the low pressure rotor velocity variations of machine 1, the low pressure of m seconds engines 2 turns before and after height reaches test point
The standard deviation (N12_std) of sub- velocity variations, when height reach test point before and after m seconds engines 1 high pressure rotor velocity variations
Standard deviation (N21_std), when height reach test point before and after m seconds engines 2 high pressure rotor velocity variations standard deviation
(N22_std).It is of course understood that number can be other for the engine of flying object, such as fly
Object may include 4 engines, in such a case, it is possible to which the parameter of other 2 engines is introduced.
After obtaining historical operating parameter, according to step S140, the disclosure can be counted the historical operating parameter
Data characteristics is obtained according to analysis, the data characteristics includes the density function of each historical operating parameter, each history run ginseng
State and correlation between number.
For example, can use the scatterplot matrices that the historical operating parameter is established between two two parameters;It is dissipated according to described
Point diagram matrix determines the data characteristics.
Fig. 4 b is please referred to, Fig. 4 b shows the signal in the scatterplot matrices established into the nearly stage using historical operating parameter
Figure.
Available from Fig. 4 b: 1) N11_std, N12_std, N21_std and N22_std are highly relevant, therefore can select
It selects and represents N12_std, N21_std and N22_std using N11_std to carry out model training (it is of course also possible to select other
One parameter is as representative);2) Head_diff is discrete parameter, and the flight analyte detection that peels off will be multiclass clustering problem;3) it removes
Other than Head_diff, the density function of other operating parameters is unimodal, approximate normal distribution, and each scatter plot have it is bright
Aobvious data converge attribute.
According to step S150, the flying object that peels off can be established according to the data characteristics and determine model.
In one example, since there are discrete parameter Head_diff in historical operating parameter, then peel off flight analyte detection
Problem is multiclass clustering problem, i.e., multiple normal flight species and the flight species that peel off are based on this, can choose corresponding multiclass
Clustering algorithm carries out the flight analyte detection that peels off, for example, DBSCAN algorithm can be selected.
Please refer to Fig. 4 c, Fig. 4 c, which is shown, determines that the parameter of model was adjusted shows to the flying object that peels off into the nearly stage
It is intended to.
Herein, for convenience, tri- history run ginsengs of Head_diff, IVV_ave and IASminusVREF are chosen
Number is analyzed, it should which explanation, the model and training method are equally applicable when multi-parameter.There are two parameters for DBSCAN algorithm
It needs to adjust, i.e. parameter eps and parameter min_samples, wherein eps indicates the radius of neighbourhood specified in DBSCAN algorithm,
Min_samples indicates the smallest sample number for including in kernel object neighborhood specified in DBSCAN algorithm.
Under the premise of avoiding over-fitting and poor fitting, parameter eps and parameter min_samples are adjusted, until
The quantity accounting of flying object of being adjusted to peel off is best between 5%~10%.
It is like before, when adjusting parameter, it can first fix a parameter and adjust another parameter, for example, please referring to
Fig. 4 c the first row fixes min_sample first to adjust min_samples, and the first width figure is in over-fitting, and third width figure is in
Poor fitting, the second width figure are normal.Then the parameter min_sample for fixing the second width figure is adjusted parameter eps, and such as second
Shown in row.Iteration repeatedly, until the quantity accounting for the flying object that is adjusted to peel off is between 5%~10%.
As illustrated in fig. 4 c, dark point (ellipse irises out part) is the flying object that peels off being marked in training process, light color
Point is the normal flight object being marked in training process.
As illustrated in fig. 4 c, in this example, when parameter eps be 0.18 or 0.22, position 5 parameter min_samples the case where
Under, the flying object that peels off into the nearly stage determines that model has preferable performance.
After determining that model training is good to the flying object that peels off, i.e., using peel off flying object determine model to need carry out from
The flying object of group's judgement is judged.
Fig. 4 d is please referred to, Fig. 4 d, which is shown, determines that model is carried out in the second test point using the flying object that peels off into the nearly stage
The schematic diagram that the flying object that peels off determines.Wherein, 0.18 is set by eps and set 5 for min_samples, it will at least one
Operating parameter (PITCH_ave, Head_diff, IASminusVREF, IVV_ave and the N11_ of a flying object (such as 480)
Std) the input flying object that peels off determines in model, and the flying object that can determine to peel off is about 43 (the dark points in Fig. 4 d),
Account for 9.03% or so of all training datas.
Fig. 4 e is please referred to, Fig. 4 e, which is shown, determines that model is carried out in third test point using the flying object that peels off into the nearly stage
The schematic diagram that the flying object that peels off determines.Wherein, 0.22 is set by eps and set 5 for min_samples, it will at least one
Operating parameter (PITCH_ave, Head_diff, IASminusVREF, IVV_ave and the N11_ of a flying object (such as 480)
Std) the input flying object that peels off determines in model, and the flying object that can determine to peel off is about 45 (the dark points in Fig. 4 e),
Account for 9.45% or so of all training datas.
Fig. 4 f is please referred to, Fig. 4 f, which is shown, determines that model is carried out in the 4th test point using the flying object that peels off into the nearly stage
The schematic diagram that the flying object that peels off determines.Wherein, 0.18 is set by eps and set 5 for min_samples, it will at least one
Operating parameter (PITCH_ave, Head_diff, IASminusVREF, IVV_ave and the N11_ of a flying object (such as 480)
Std) the input flying object that peels off determines in model, and the flying object that can determine to peel off is about 40 (the dark points in Fig. 4 f),
Account for 8.40% or so of all training datas.
For into the nearly stage, the number to the predetermined detection point of the flying object that peels off detected and selected is 3, therefore, In
Obtain the second test point, third test point, the 4th test point the flying object that peels off result when, can be according to " determining described pre-
Determine the estimation in the landing stage under N number of default test point to peel off flying object;By be confirmed as estimation peel off flying object number not
Flying object less than M is determined as the flying object that peels off " further determine that the flying object that peels off.For example, the second detection can be integrated
The flying object that peels off that point (1000 feet), third test point (500 feet) and the 4th test point (300 feet) detect respectively,
Further screening persistently peels off flying object.
Fig. 4 g is please referred to, Fig. 4 g shows the signal that the flying object that peels off is determined in conjunction with multiple predetermined detection points into the nearly stage
Figure.
When the minimum detection number for setting the flying object that persistently peels off is 3 (M), 12 flying objects that persistently peel off are marked altogether,
It is marked in figure 4g with dark color point.These flying objects are controlled not over flight by the takeoff and landing performance adjustment and recovery of aircraft, these
Although flying object not necessarily triggers alarm, relative community performance, the risk that problem occurs for the flying object that persistently peels off is higher, needs
It pays close attention to.
For landing flat float the stage:
In a kind of possible embodiment, the predetermined landing stage be landing it is flat float the stage in the case where, it is described
Default test point includes the 5th test point, the 6th test point, the 7th test point, wherein the 5th test point be flying object from
Position where when pitching rule is revised as evening up rule, the 6th test point are as defined in flying object reaches under stable condition
Position where when flare out altitude, the 7th test point, which receives to propaganda directed to communicate automatically for flying object, reminds pilot to withdraw thrust hand
Position where when handle.
In a kind of possible embodiment, the height of the 5th test point is 50 feet, the 6th test point
Height is 30 feet, and the height of the 7th test point is 20 feet.
In a kind of possible embodiment, the predetermined landing stage be landing it is flat float the stage in the case where, it is described
Operating parameter may include Inertial Vertica1 Speed mean value (IVV_ave), pitch angle mean value (PITCH_ave), air speed mean value (IAS_
Ave), vertically overload mean value (VRTG_ave), ground distance (DIST_LD) and be grounded time (TIME_LD) any one and
It is a variety of, wherein the Inertial Vertica1 Speed mean value indicates that flying object height reaches the default each second k seconds inertia in test point front and back and hangs down
The average value of straight speed (IVV), the pitch angle mean value indicate before and after flying object height reaches default test point each second k seconds
The average value of pitch angle, the air speed mean value indicate that flying object height reaches each second k seconds air speed in default test point front and back
(IAS) average value, the vertical overload mean value indicate before and after flying object height reaches default test point each second k seconds vertical
The average value of VRTG is overloaded, the horizontal distance of grounding point to pick-up point is preset described in the ground distance, the ground connection time indicates to fly
Time of the row object from the default grounding point to pick-up point, wherein k > 0.
In a kind of possible embodiment, the specific size of k can determines according to actual conditions, in one example, k
It can be 5s.
The example in the operating parameter in landing flat stage of floaing is presented above, it should be appreciated that, those skilled in the art
It can increase according to the actual situation, reduce operating parameter, in this regard, the disclosure is without limitation.
The determination of model training and the flying object that peels off in the landing flat stage of floaing is illustrated below.
Fig. 5 a is please referred to, Fig. 5 a shows the historical operating parameter in the same model flying object of the flat stage acquisition of floaing of landing
Schematic diagram.
According to step S130, it is winged that the disclosure can obtain the same model as shown in Figure 5 a in the flat stage acquisition of floaing of landing
The historical operating parameter of row object.
As shown in Figure 5 a, in one example, the disclosure has selected 28 flying objects (number is No-X0001~X0028)
Historical operating parameter, in a kind of possible embodiment, the predetermined landing stage be landing it is flat float the stage the case where
Under, the operating parameter includes Inertial Vertica1 Speed mean value (IVV_ave), pitch angle mean value (PITCH_ave), air speed mean value
(IAS_ave), mean value (VRTG_ave), ground distance (DIST_LD) and ground connection time (TIME_LD) are vertically overloaded.
After obtaining historical operating parameter, according to step S140, the disclosure can be counted the historical operating parameter
Data characteristics is obtained according to analysis, the data characteristics includes the density function of each historical operating parameter, each history run ginseng
State and correlation between number.
For example, can use the scatterplot matrices that the historical operating parameter is established between two two parameters;
The data characteristics is determined according to the scatterplot matrices.
Fig. 5 b is please referred to, Figure 5b shows that utilize the scatterplot matrices of historical operating parameter foundation in the landing flat stage of floaing
Schematic diagram.
From Fig. 5 b it can be seen that 1) in Fig. 5 b, the density function on diagonal line is provided with single-peak response, similar normal state point
Cloth;2) there is apparent data to converge characteristic between parameter and parameter, and without discrete parameter;3) DIST_LD and TIME_LD high
Degree is related, thus can choose DIST_LD represent TIME_LD be trained (when the flying object that peels off later judges, can also be with
TIME_LD is represented with DIST_LD).
According to step S150, the flying object that peels off can be established according to the data characteristics and determine model.
In one example, discrete parameter is not present in historical operating parameter, then the flying object test problems that peel off can be seen
At being single class clustering problem, i.e., normal flight species and the flight species that peel off can choose corresponding single class cluster and calculated based on this
Method carries out the flight analyte detection that peels off, such as can choose One-class SVM algorithm.
Fig. 5 c is please referred to, Fig. 5 c, which is shown, determines that the parameter of model is adjusted to the flying object that peels off in landing flat stage of floaing
Schematic diagram.
Herein, for convenience, it chooses IVV_ave and two historical operating parameters progress flying objects of DIST_LD is different
It often tests and analyzes, it should which explanation, the model and training method are equally applicable when multi-parameter.The disclosure illustratively selects
One-Class SVM algorithm carries out clustering learning, wherein kernel function selects radial basis function (RBF), and to two cores of algorithm
The heart parameter nu and gamma carry out parameter adjustment, and nu is used to provide the training error coboundary of SVM algorithm, and gamma is used to adjust diameter
To the scale parameter of basic function, under the premise of avoiding over-fitting and poor fitting, until the quantity for the flying object that is adjusted to peel off accounts for
Than being most preferably, shown in result figure 5c between 5%~10%.
It is that the flying object that peels off determines the flying object that peels off that model marks as described in Fig. 5 c, inside line of demarcation, outside line of demarcation
It is that the flying object that peels off determines the normal point that model marks, line of demarcation is that the flying object that peels off determines the classification side that model learning obtains
Boundary.It can be seen that parameter nu is bigger, classification boundaries are more like center convergence;Parameter gamma can be used to adjust the shape of classification boundaries
Shape, the parameter is bigger, and classification boundaries are more bent, but the parameter is excessive, over-fitting easily occurs (such as legend institute in the lower right corner in Fig. 5 c
Show).It is that over-fitting and poor fitting can all influence abnormality detection as a result, the disclosure is avoided in parameter adjustment.
It can be seen that in the case that parameter nu is 0.05, parameter gamma is 5 from Fig. 5 c, the landing flat stage of floaing peels off
Flying object determines that model has preferable performance.
After determining that model training is good to the flying object that peels off, i.e., using peel off flying object determine model to need carry out from
The flying object of group's judgement is judged.
Fig. 5 d is please referred to, Fig. 5 d is shown determines model in the 5th test point using the flying object that peels off in landing flat stage of floaing
Carry out the determining schematic diagram of flying object that peels off.Wherein, in this example, 0.05 is set by parameter nu, parameter ganmma is set
Be set to 5, select at least one flying object at the 5th test point (50 feet) operating parameter (IVV_ave, PITCH_ave,
IAS_ave, VRTG_ave and DIST_LD) the input flying object that peels off determines in model, the flying object that can determine to peel off is about
27 (the dark point in Fig. 5 d), 5.67% or so of Zhan Suoyou training data.
Fig. 5 e is please referred to, Fig. 5 e is shown determines model in the 6th test point using the flying object that peels off in landing flat stage of floaing
Carry out the determining schematic diagram of flying object that peels off.Wherein, in this example, 0.05 is set by parameter nu, parameter ganmma is set
Be set to 5, select at least one flying object at the 6th test point (30 feet) operating parameter (IVV_ave, PITCH_ave,
IAS_ave, VRTG_ave and DIST_LD) the input flying object that peels off determines in model, the flying object that can determine to peel off is about
25 (the dark point in Fig. 5 e), 5.25% or so of Zhan Suoyou training data.
Fig. 5 f is please referred to, Fig. 5 f is shown determines model in the 7th test point using the flying object that peels off in landing flat stage of floaing
Carry out the determining schematic diagram of flying object that peels off.Wherein, in this example, 0.05 is set by parameter nu, parameter ganmma is set
Be set to 5, select at least one flying object at the 7th test point (20 feet) operating parameter (IVV_ave, PITCH_ave,
IAS_ave, VRTG_ave and DIST_LD) the input flying object that peels off determines in model, the flying object that can determine to peel off is about
27 (the dark point in Fig. 5 f), 5.67% or so of Zhan Suoyou training data.
It should be noted that can choose during the flying object that peels off in each stage to the landing stage determines
Except the predetermined detection point that the disclosure enumerates, a kind of example is given below.
Fig. 5 g is please referred to, Fig. 5 g is shown determines model in the 8th test point using the flying object that peels off in landing flat stage of floaing
Carry out the determining schematic diagram of flying object that peels off.Wherein, in this example, 0.05 is set by parameter nu, parameter ganmma is set
Be set to 5, select at least one flying object at the 8th test point (10 feet) operating parameter (IVV_ave, PITCH_ave,
IAS_ave, VRTG_ave and DIST_LD) the input flying object that peels off determines in model, the flying object that can determine to peel off is about
24 (the dark point in Fig. 5 g), 5.04% or so of Zhan Suoyou training data.
For landing for the flat stage of floaing, the number to the predetermined detection point of the flying object that peels off detected and selected is 4, because
This, obtain the 5th test point, the 6th test point, the 7th test point, the 8th test point the flying object that peels off result when, can be with
According to " determining that the estimation in the predetermined landing stage under N number of default test point peels off flying object;It will be confirmed as estimation to peel off
Flying object of the number of flying object not less than M is determined as the flying object that peels off " further determine that the flying object that peels off.For example, can
With comprehensive 5th test point (50 feet), the 6th test point (30 feet), the 7th test point (20 feet), the 8th test point (10
Foot) flying object that peels off that detects respectively, further screening persistently peels off flying object.
Fig. 5 h is please referred to, Fig. 5 h shows the determining flying object that peels off of multiple predetermined detection points in conjunction with the landing flat stage of floaing
Schematic diagram.
When the minimum detection number for setting the flying object that persistently peels off is 4 (M), 8 flying objects that persistently peel off, In are marked altogether
It is marked in Fig. 5 h with dark color point.Not over flight control by the takeoff and landing performance adjustment and recovery of aircraft, these fly these flying objects
Although row object not necessarily triggers alarm, relative community performance, the risk that problem occurs for the flying object that persistently peels off is higher, needs
It pays close attention to.
It according to described above it is found that the flying object that peels off described in the disclosure determines that method is entirely data-driven, and is complete
Industry data, multi-parameter are analyzed simultaneously, avoid the dependence of subjective experience, cover the data characteristics of the whole industry, contain ginseng
Several correlation properties.This method is implemented without priori threshold value, and peel off flying object and judgement side are found completely from data
Boundary, it is possible to reduce influence of the subjective experience to management, and the flying object that takeoff and landing performance peels off can be objectively extracted, aviation is public
Department, which can be absorbed in, peels off flying object to improve security management services level, and pilot can use these information to improve them
Airmanship.
Referring to Fig. 6, Fig. 6 shows the block diagram of the flying object determining device that peels off according to one embodiment of the disclosure.Such as
Shown in Fig. 6, described device includes: the first acquisition module 10, is reached for obtaining at least one flying object in the predetermined landing stage
Operating parameter when default test point;Determining module 20 is connected to the acquisition module 10, for using the operating parameter and
The flying object that peels off corresponding with predetermined landing stage determines that model determines the flight that peels off at least one described flying object
Object.
By apparatus above, at least one available flying object of the disclosure reaches default test point in the predetermined landing stage
When operating parameter, determine that model determines using the operating parameter and the flying object that peels off corresponding with predetermined landing stage
The flying object that peels off at least one described flying object.The disclosure can quickly and accurately determine at least one flying object
Peel off flying object, also, the disclosure is to carry out comprehensive descision to the flying object that peels off based on whole industry QAR data and multiple parameters,
Avoid the dependence to subjective experience.
Using device described in the disclosure determined at least one flying object peel off flying object in the case where, indicate
To preset the takeoff and landing performance deviation near test point larger at this for takeoff and landing performance of the flying object near default test point and group,
It needs to be adjusted.After obtaining result, control centre can be sent result to and referred to, or police can be issued
It accuses, to remind staff to pay attention to averting risks.
In a kind of possible embodiment, the determining module includes: the first determining submodule, described pre- for determining
Determine the estimation in the landing stage under N number of default test point to peel off flying object;Second determines submodule, is connected to described first and determines
Submodule is determined as the flying object that peels off not less than the flying object of M for will be confirmed as the peel off number of flying object of estimation,
Wherein, M, N are natural number, M≤N.
Referring to Fig. 7, Fig. 7 shows the block diagram of the flying object determining device that peels off according to one embodiment of the disclosure.In
In a kind of possible embodiment, as shown in fig. 7, described device can also include: the second acquisition module 30, it is multiple for obtaining
Flying object reaches the historical operating parameter of default test point in the predetermined landing stage;Analysis module 40 is connected to described second and obtains
Modulus block 30 is analyzed to obtain data characteristics for carrying out data to the historical operating parameter, and the data characteristics includes each
State and correlation between the density function of historical operating parameter, each historical operating parameter;Module 50 is established, institute is connected to
Analysis module 40 and the first acquisition module 10 are stated, determines model for establishing the flying object that peels off according to the data characteristics.
It is described that historical operating parameter progress data are analyzed to obtain data spy in a kind of possible embodiment
Sign, comprising: establish the scatterplot matrices between two two parameters using the historical operating parameter;It is true according to the scatterplot matrices
The fixed data characteristics.In a kind of possible embodiment, the flying object that peels off is established according to the data characteristics and determines model,
It include: there are historical operating parameters to be rendered as discrete state, or, there are the density functions of historical operating parameter, and multimodal spy is presented
Property in the case where, determine described in peel off flying object determine model be multiclass Clustering Model;Or it is in all historical operating parameters
In the case that single-peak response is presented in the density function of the poly- state of cash and all historical operating parameters, peel off flight described in determination
Object determines model for single class Clustering Model.
In a kind of possible embodiment, there are in the case where correlation between multiple historical operating parameters, utilize
Other historical operating parameters of one of multiple historical operating parameters with correlation and not correlation are as training
Data.
It is described that model is determined according to the data characteristics foundation flying object that peels off in a kind of possible embodiment, also
Include: under the premise of avoiding over-fitting and poor fitting, using the training data to the flying object that peels off determine model into
Row training;The model parameter of model, which is adjusted, to be determined to the flying object that peels off, when the flying object that peels off determined in training
Number when accounting for all flying object numbers and reaching the first ratio, determine described in the flying object that peels off determine the final mask ginseng of model
Number, and the flying object that peels off after being trained according to the final mask parameter determines model.
In a kind of possible embodiment, first ratio is between 5% to 10%.
It is described default in the case where the predetermined landing stage is takeoff phase in a kind of possible embodiment
Test point includes the first test point, and first test point is the position of the moment of taking off of flying object, the operating parameter
Including distance of taking off, flying speed, the liftoff elevation angle, liftoff elevation angle change rate mean value, liftoff elevation angle change rate standard deviation, it is liftoff hang down
Any one or more of straight speed, flying speed change rate mean value, flying speed change rate standard deviation, wherein described to take off
Distance indicates that flying object flies to horizontal distance when flying object reaches default safe altitude from the outset, and the flying speed indicates
In the horizontal velocity of flying object main wheel liftoff instant flying object, the liftoff elevation angle indicates to fly in flying object main wheel liftoff instant
The elevation angle of object, the liftoff elevation angle change rate mean value indicate that each second elevation angle changes equal in flying object main wheel liftoff front and back n seconds
Value, the liftoff elevation angle change rate standard deviation indicate the standard deviation of each second liftoff front and back n seconds elevation angle of flying object main wheel variation, institute
State vertical speed of the liftoff vertical speed expression in flying object main wheel liftoff instant flying object, the flying speed change rate mean value
Indicate the mean value of each second liftoff front and back n seconds vertical speed of flying object main wheel variation, the flying speed change rate standard deviation table
Show the standard deviation of each second liftoff front and back n seconds vertical speed of flying object main wheel variation, wherein n > 0.
It is described default in the case where the predetermined landing stage is into the nearly stage in a kind of possible embodiment
Test point includes the second test point, third test point and the 4th test point, and second test point is that flying object reaches instrument supply gas
Position where when as the stabilization under condition IMC into nearly detection height, second test point is that flying object reaches visual meteorological
Position where when stabilization under condition VMC is into nearly detection height, when the third test point is that flying object is in five side of airport
Stabilization into nearly detection height when where position.
In a kind of possible embodiment, the height of second test point is 1000 feet, the third test point
Height be 500 feet, the height of the 4th test point is 300 feet.
In a kind of possible embodiment, in the case where the predetermined landing stage is into the nearly stage, the operation
Parameter includes pitch angle mean value, course mean value, relative velocity mean value, vertical speed mean value, each engine low pressure rotor standard
Any one or more of poor, each engine high pressure rotor standard deviation, wherein the pitch angle mean value indicates flying object height
The average value of each second m seconds pitch angle in default test point front and back is reached, it is pre- that the course mean value indicates that flying object height reaches
If the average value of each second m seconds course variation before and after test point, it is default that the relative velocity mean value indicates that flying object height reaches
Each second m seconds air speed subtracts the average value of reference velocity before and after test point, and the vertical speed mean value indicates that flying object height arrives
The average value of each second m seconds Inertial Vertica1 Speed before and after up to default test point, the engine low pressure rotor standard deviation indicate to fly
The standard deviation of row object low pressure rotor velocity variations of each second m seconds engine before and after height reaches and presets test point, the hair
Motivation high pressure rotor standard deviation indicates that the low pressure rotor speed of each second m seconds engine before and after height reaches default test point becomes
The standard deviation of change, wherein m > 0.
In a kind of possible embodiment, the predetermined landing stage be landing it is flat float the stage in the case where, it is described
Default test point includes the 5th test point, the 6th test point, the 7th test point, wherein the 5th test point be flying object from
Position where when pitching rule is revised as evening up rule, the 6th test point are as defined in flying object reaches under stable condition
Position where when flare out altitude, the 7th test point, which receives to propaganda directed to communicate automatically for flying object, reminds pilot to withdraw thrust hand
Position where when handle.
In a kind of possible embodiment, the height of the 5th test point is 50 feet, the 6th test point
Height is 30 feet, and the height of the 7th test point is 20 feet.
In a kind of possible embodiment, the predetermined landing stage be landing it is flat float the stage in the case where, it is described
Operating parameter includes Inertial Vertica1 Speed mean value, pitch angle mean value, air speed mean value, vertically overloads mean value, ground distance and ground connection
Time any one and it is a variety of, wherein before the Inertial Vertica1 Speed mean value indicates that flying object height reaches default test point
The average value of each second k seconds Inertial Vertica1 Speed afterwards, before the pitch angle mean value indicates that flying object height reaches default test point
The average value of each second k seconds pitch angle afterwards, the air speed mean value indicate that flying object height reaches default test point front and back k seconds often
The average value of one second air speed (IAS), the vertical overload mean value indicate that flying object height reaches default test point front and back k seconds often
The average value of one second vertical overload VRTG presets the horizontal distance of grounding point to pick-up point, ground connection described in the ground distance
Time indicates time of the flying object from the default grounding point to pick-up point, wherein k > 0.
It according to described above it is found that the flying object determining device that peels off described in the disclosure is entirely data-driven, and is complete
Industry data, multi-parameter are analyzed simultaneously, avoid the dependence of subjective experience, cover the data characteristics of the whole industry, contain ginseng
Several correlation properties.This method is implemented without priori threshold value, and peel off flying object and judgement side are found completely from data
Boundary, it is possible to reduce influence of the subjective experience to management, and the flying object that takeoff and landing performance peels off can be objectively extracted, aviation is public
Department, which can be absorbed in, peels off flying object to improve security management services level, and pilot can use these information to improve them
Airmanship.
The presently disclosed embodiments is described above, above description is exemplary, and non-exclusive, and
It is not limited to disclosed each embodiment.Without departing from the scope and spirit of illustrated each embodiment, for this skill
Many modifications and changes are obvious for the those of ordinary skill in art field.The selection of term used herein, purport
In the principle, practical application or improvement to the technology in market for best explaining each embodiment, or make the art
Other those of ordinary skill can understand each embodiment disclosed herein.
Claims (13)
1. one kind peels off, flying object determines method, which is characterized in that the described method includes:
Obtain operating parameter of at least one flying object when reaching default test point in the predetermined landing stage;
Using the operating parameter and the flying object that peels off corresponding with predetermined landing stage determine model determine it is described at least
The flying object that peels off in one flying object.
2. the method according to claim 1, wherein it is described using the operating parameter and with the predetermined landing
The stage corresponding flying object that peels off determines that model determines the flying object that peels off at least one described flying object, comprising:
Determine that the estimation in the predetermined landing stage under N number of default test point peels off flying object;
It will be confirmed as the peel off number of flying object of estimation and be determined as the flying object that peels off not less than the flying object of M,
Wherein, M, N are natural number, M≤N.
3. the method according to claim 1, wherein the method also includes:
Obtain the historical operating parameter that multiple flying objects reach default test point in the predetermined landing stage;
It carries out data to the historical operating parameter to analyze to obtain data characteristics, the data characteristics includes each history run ginseng
State and correlation between several density function, each historical operating parameter;
The flying object that peels off, which is established, according to the data characteristics determines model.
4. according to the method described in claim 3, it is characterized in that, described analyze historical operating parameter progress data
To data characteristics, comprising:
The scatterplot matrices between two two parameters are established using the historical operating parameter;
The data characteristics is determined according to the scatterplot matrices.
5. the method according to claim 3 or 4, which is characterized in that it is true to establish the flying object that peels off according to the data characteristics
Cover half type, comprising:
There are historical operating parameters to be rendered as discrete state, or, there are the density functions of historical operating parameter, and multimodal spy is presented
Property in the case where, determine described in peel off flying object determine model be multiclass Clustering Model;Or
Unimodal spy is presented in the density function that convergence state and all historical operating parameters is presented in all historical operating parameters
Property in the case where, determine described in the flying object that peels off determine model for single class Clustering Model.
6. according to the method described in claim 5, it is characterized in that, there are the feelings of correlation between multiple historical operating parameters
Under condition, one of multiple historical operating parameters with correlation and not no other historical operating parameters of correlation are utilized
As training data.
7. according to the method described in claim 6, it is characterized in that, described true according to the data characteristics foundation flying object that peels off
Cover half type, further includes:
Under the premise of avoiding over-fitting and poor fitting, model, which carries out, to be determined to the flying object that peels off using the training data
Training;
The model parameter of model, which is adjusted, to be determined to the flying object that peels off, when the number of the flying object that peels off determined in training
When mesh accounts for all flying object numbers and reaches the first ratio, determine described in the flying object that peels off determine the final mask parameter of model, and
The flying object that peels off after being trained according to the final mask parameter determines model.
8. the method according to claim 1, wherein the case where the predetermined landing stage is takeoff phase
Under, the default test point includes the first test point, and first test point is the position of the moment of taking off of flying object, institute
Stating operating parameter includes take off distance, flying speed, the liftoff elevation angle, liftoff elevation angle change rate mean value, liftoff elevation angle change rate mark
Any one or more of quasi- poor, liftoff vertical speed, flying speed change rate mean value, flying speed change rate standard deviation,
In, the distance of taking off indicates that flying object flies to horizontal distance when flying object reaches default safe altitude from the outset, described
Flying speed indicates the horizontal velocity in flying object main wheel liftoff instant flying object, and the liftoff elevation angle is indicated in flying object main wheel
The elevation angle of liftoff instant flying object, the liftoff elevation angle change rate mean value indicate each second in flying object main wheel liftoff front and back n seconds
The mean value of elevation angle variation, the liftoff elevation angle change rate standard deviation indicate that each second liftoff front and back n seconds elevation angle of flying object main wheel becomes
The standard deviation of change, the liftoff vertical speed indicates the vertical speed in flying object main wheel liftoff instant flying object, described liftoff
Percentage speed variation mean value indicates the mean value of each second liftoff front and back n seconds vertical speed of flying object main wheel variation, the flying speed
Change rate standard deviation indicates the standard deviation of each second liftoff front and back n seconds vertical speed of flying object main wheel variation, wherein n > 0.
9. the method according to claim 1, wherein the case where the predetermined landing stage is into the nearly stage
Under, the default test point includes the second test point, third test point and the 4th test point, and second test point is flying object
Position where when reaching stabilization under instrument meteoro logical condition, IMC IMC into nearly detection height, second test point reach for flying object
Position where when stabilization under to visual meteorological condition (VMC) VMC is into nearly detection height, the third test point are in for flying object
Position where when stabilization when five side of airport is into nearly detection height.
10. the method according to claim 1, wherein the case where the predetermined landing stage is into the nearly stage
Under, the operating parameter includes that pitch angle mean value, course mean value, relative velocity mean value, vertical speed mean value, each engine are low
Press any one or more of rotor standard deviation, each engine high pressure rotor standard deviation, wherein the pitch angle mean value indicates
Flying object height reaches the average value of each second m seconds pitch angle in default test point front and back, and the course mean value indicates flying object
Height reaches the average value of each second m seconds course variation before and after default test point, and the relative velocity mean value indicates that flying object is high
Each second m seconds air speed subtracts the average value of reference velocity before and after degree reaches default test point, and the vertical speed mean value indicates to fly
Row object height reaches the average value of each second m seconds Inertial Vertica1 Speed before and after default test point, the engine low pressure rotor mark
Quasi- difference indicates the standard of flying object low pressure rotor velocity variations of each second m seconds engine before and after highly reaching default test point
Difference, the engine high pressure rotor standard deviation indicate the low pressure of each second m seconds engine before and after height reaches default test point
The standard deviation of spinner velocity variation, wherein m > 0.
11. floaing the stage the method according to claim 1, wherein being that landing is flat in the predetermined landing stage
In the case of, the default test point includes the 5th test point, the 6th test point, the 7th test point, wherein the 5th test point
Position where when being revised as evening up rule from pitching rule for flying object, the 6th test point are that flying object reaches stablizing bar
Under part when defined flare out altitude where position, the 7th test point is that flying object receives and propagandas directed to communicate automatically and reminds pilot
Position where when withdrawal thrust handle.
12. floaing the stage the method according to claim 1, wherein being that landing is flat in the predetermined landing stage
In the case of, the operating parameter includes Inertial Vertica1 Speed mean value, pitch angle mean value, air speed mean value, vertically overloads mean value, ground connection
Distance and ground connection time any one and it is a variety of, wherein it is pre- that the Inertial Vertica1 Speed mean value indicates that flying object height reaches
If the average value of each second k seconds Inertial Vertica1 Speed before and after test point, it is pre- that the pitch angle mean value indicates that flying object height reaches
If the average value of each second k seconds pitch angle before and after test point, the air speed mean value indicates that flying object height reaches default detection
The average value of each second k seconds air speed (IAS) in point front and back, the vertical overload mean value indicate that flying object height reaches default detection
The average value of each second k seconds vertical overload VRTG in front and back is put, the water of grounding point to pick-up point is preset described in the ground distance
Flat distance, ground connection time indicate time of the flying object from the default grounding point to pick-up point, wherein k > 0.
The flying object determining device 13. one kind peels off, which is characterized in that described device includes:
First obtains module, for obtaining operation ginseng of at least one flying object when reaching default test point in the predetermined landing stage
Number;
Determining module is connected to the acquisition module, for utilizing the operating parameter and corresponding with the predetermined landing stage
The flying object that peels off determine that model determines the flying object that peels off at least one described flying object.
Priority Applications (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910700625.1A CN110390177B (en) | 2019-07-31 | 2019-07-31 | Method and device for determining outlier flying object |
Applications Claiming Priority (1)
Application Number | Priority Date | Filing Date | Title |
---|---|---|---|
CN201910700625.1A CN110390177B (en) | 2019-07-31 | 2019-07-31 | Method and device for determining outlier flying object |
Publications (2)
Publication Number | Publication Date |
---|---|
CN110390177A true CN110390177A (en) | 2019-10-29 |
CN110390177B CN110390177B (en) | 2023-06-16 |
Family
ID=68288108
Family Applications (1)
Application Number | Title | Priority Date | Filing Date |
---|---|---|---|
CN201910700625.1A Active CN110390177B (en) | 2019-07-31 | 2019-07-31 | Method and device for determining outlier flying object |
Country Status (1)
Country | Link |
---|---|
CN (1) | CN110390177B (en) |
Cited By (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111477036A (en) * | 2020-04-08 | 2020-07-31 | 中国电子科技集团公司第二十八研究所 | Method for detecting aircraft height abnormity of air traffic control automation system |
CN112200052A (en) * | 2020-09-30 | 2021-01-08 | 驭势(上海)汽车科技有限公司 | Trajectory deviation identification and vehicle running analysis method, device, equipment and medium |
CN114004292A (en) * | 2021-10-29 | 2022-02-01 | 重庆大学 | Pilot flat-floating ejector rod behavior analysis method based on flight parameter data unsupervised clustering |
Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN107045638A (en) * | 2016-12-30 | 2017-08-15 | 中国民航管理干部学院 | A kind of flight safety affair analytical method based on context-aware model |
US20170301247A1 (en) * | 2016-04-19 | 2017-10-19 | George Mason University | Method And Apparatus For Probabilistic Alerting Of Aircraft Unstabilized Approaches Using Big Data |
CN110033194A (en) * | 2019-04-17 | 2019-07-19 | 中国民航科学技术研究院 | A kind of aircraft is unstable into nearly detection and evaluation method and electronic equipment |
-
2019
- 2019-07-31 CN CN201910700625.1A patent/CN110390177B/en active Active
Patent Citations (3)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
US20170301247A1 (en) * | 2016-04-19 | 2017-10-19 | George Mason University | Method And Apparatus For Probabilistic Alerting Of Aircraft Unstabilized Approaches Using Big Data |
CN107045638A (en) * | 2016-12-30 | 2017-08-15 | 中国民航管理干部学院 | A kind of flight safety affair analytical method based on context-aware model |
CN110033194A (en) * | 2019-04-17 | 2019-07-19 | 中国民航科学技术研究院 | A kind of aircraft is unstable into nearly detection and evaluation method and electronic equipment |
Non-Patent Citations (2)
Title |
---|
李晓明等: "下降进近阶段飞行安全分析与建模", 《信息技术》 * |
杨慧等: "基于聚类和拟合的QAR数据离群点检测算法", 《计算机工程与设计》 * |
Cited By (5)
Publication number | Priority date | Publication date | Assignee | Title |
---|---|---|---|---|
CN111477036A (en) * | 2020-04-08 | 2020-07-31 | 中国电子科技集团公司第二十八研究所 | Method for detecting aircraft height abnormity of air traffic control automation system |
CN112200052A (en) * | 2020-09-30 | 2021-01-08 | 驭势(上海)汽车科技有限公司 | Trajectory deviation identification and vehicle running analysis method, device, equipment and medium |
CN112200052B (en) * | 2020-09-30 | 2023-06-09 | 驭势(上海)汽车科技有限公司 | Track deviation recognition and vehicle running analysis method, device, equipment and medium |
CN114004292A (en) * | 2021-10-29 | 2022-02-01 | 重庆大学 | Pilot flat-floating ejector rod behavior analysis method based on flight parameter data unsupervised clustering |
CN114004292B (en) * | 2021-10-29 | 2023-02-03 | 重庆大学 | Pilot flat-floating ejector rod behavior analysis method based on flight parameter data unsupervised clustering |
Also Published As
Publication number | Publication date |
---|---|
CN110390177B (en) | 2023-06-16 |
Similar Documents
Publication | Publication Date | Title |
---|---|---|
US11610495B2 (en) | Unmanned aerial vehicle authorization and geofence envelope determination | |
US10643480B2 (en) | Method and apparatus for probabilistic alerting of aircraft unstabilized approaches using big data | |
CN110390177A (en) | The flying object that peels off determines method and device | |
US7586422B2 (en) | Determination of runway landing conditions | |
FR3046225B1 (en) | DISPLAY OF WEATHER DATA IN AN AIRCRAFT | |
US8560150B1 (en) | Methods and systems for landing decision point | |
EP3187826B1 (en) | Display of meteorological data in an aircraft | |
RU2302359C2 (en) | Method and device for detection of impairment of flying vehicle characteristics | |
US20090267798A1 (en) | The communication of landing conditions | |
US8780091B2 (en) | Methods and systems for controlling an information display | |
EP2816432A1 (en) | Flight deck display systems and methods for visually indicating low speed change conditions during takeoff and landing | |
CN110473431A (en) | A kind of conflict prediction method under the height limitation of approach control airspace | |
CN110275541A (en) | The optimization track of noise is improved using automatic takeoff | |
JP2009192262A (en) | System and method for predicting air turbulence | |
CN105807087A (en) | Method and device for determining the wind speed to be taken into account in order to optimize the takeoff weight of an aircraft | |
US7206674B1 (en) | Information display system for atypical flight phase | |
US8224506B2 (en) | Method and device for determining a maximum stabilization height in the final flight phase of an airplane | |
Rustenburg et al. | An evaluation of methods to separate maneuver and gust load factors from measured acceleration time histories | |
FR3038751A1 (en) | METHOD FOR INTEGRATING A CONSTRAINED ROAD OPTIMIZATION APPLICATION IN AN OPEN ARCHITECTURE AIRCRAFT SYSTEM OF CLIENT-TYPE SERVER | |
CN116415818B (en) | Method and system for confirming risk points in aircraft approach stage based on clustering algorithm | |
CN108198462B (en) | A kind of full airspace aircraft wake experience risk warning system implementation method | |
US10876857B1 (en) | Wake vortex display system | |
FR3079336A1 (en) | AIRCRAFT OPERATIONAL FLIGHT PLAN SYSTEM AND ASSOCIATED METHOD | |
Payan et al. | Improvement of rotorcraft safety metrics using performance models and data integration | |
CN111504341A (en) | Helicopter flight state identification method |
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 |