CN108595756A - The method and device of big envelope curve flight Interference Estimation - Google Patents
The method and device of big envelope curve flight Interference Estimation Download PDFInfo
- Publication number
- CN108595756A CN108595756A CN201810233100.7A CN201810233100A CN108595756A CN 108595756 A CN108595756 A CN 108595756A CN 201810233100 A CN201810233100 A CN 201810233100A CN 108595756 A CN108595756 A CN 108595756A
- Authority
- CN
- China
- Prior art keywords
- elastic
- aircraft
- signal
- large envelope
- flight
- 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
- 238000000034 method Methods 0.000 title claims abstract description 60
- 238000004458 analytical method Methods 0.000 claims abstract description 22
- 238000010586 diagram Methods 0.000 claims abstract description 21
- 238000005516 engineering process Methods 0.000 claims abstract description 5
- 238000012545 processing Methods 0.000 claims description 10
- 238000001914 filtration Methods 0.000 claims description 9
- 230000004927 fusion Effects 0.000 claims description 9
- 238000010183 spectrum analysis Methods 0.000 claims description 7
- RZVHIXYEVGDQDX-UHFFFAOYSA-N 9,10-anthraquinone Chemical compound C1=CC=C2C(=O)C3=CC=CC=C3C(=O)C2=C1 RZVHIXYEVGDQDX-UHFFFAOYSA-N 0.000 claims description 5
- 230000003044 adaptive effect Effects 0.000 claims description 5
- 238000012937 correction Methods 0.000 claims description 4
- 238000003062 neural network model Methods 0.000 claims description 4
- 230000007246 mechanism Effects 0.000 abstract description 20
- 238000013135 deep learning Methods 0.000 abstract description 12
- 238000013527 convolutional neural network Methods 0.000 abstract description 3
- 230000010355 oscillation Effects 0.000 abstract description 2
- 230000008878 coupling Effects 0.000 description 20
- 238000010168 coupling process Methods 0.000 description 20
- 238000005859 coupling reaction Methods 0.000 description 20
- 230000004044 response Effects 0.000 description 9
- 230000000694 effects Effects 0.000 description 7
- 238000004364 calculation method Methods 0.000 description 6
- 238000006880 cross-coupling reaction Methods 0.000 description 6
- 230000008569 process Effects 0.000 description 6
- 241000272517 Anseriformes Species 0.000 description 5
- 230000008859 change Effects 0.000 description 5
- 230000001808 coupling effect Effects 0.000 description 5
- 238000013461 design Methods 0.000 description 5
- 238000000605 extraction Methods 0.000 description 4
- 239000000463 material Substances 0.000 description 4
- 238000005070 sampling Methods 0.000 description 4
- 230000009286 beneficial effect Effects 0.000 description 3
- 238000004422 calculation algorithm Methods 0.000 description 3
- 230000000737 periodic effect Effects 0.000 description 3
- 230000001133 acceleration Effects 0.000 description 2
- 238000013528 artificial neural network Methods 0.000 description 2
- 238000013016 damping Methods 0.000 description 2
- 229920001971 elastomer Polymers 0.000 description 2
- 239000000806 elastomer Substances 0.000 description 2
- 239000000446 fuel Substances 0.000 description 2
- 238000002955 isolation Methods 0.000 description 2
- 238000010801 machine learning Methods 0.000 description 2
- 238000005259 measurement Methods 0.000 description 2
- 238000011160 research Methods 0.000 description 2
- 238000006467 substitution reaction Methods 0.000 description 2
- QKYZQKJXAKNEPA-UHFFFAOYSA-N 3-hydroxy-5-methyl-1-phenylthieno[2,3-d]pyrimidine-2,4-dione Chemical compound O=C1N(O)C(=O)C=2C(C)=CSC=2N1C1=CC=CC=C1 QKYZQKJXAKNEPA-UHFFFAOYSA-N 0.000 description 1
- 101000606537 Homo sapiens Receptor-type tyrosine-protein phosphatase delta Proteins 0.000 description 1
- 102100039666 Receptor-type tyrosine-protein phosphatase delta Human genes 0.000 description 1
- QVGXLLKOCUKJST-UHFFFAOYSA-N atomic oxygen Chemical compound [O] QVGXLLKOCUKJST-UHFFFAOYSA-N 0.000 description 1
- 230000002238 attenuated effect Effects 0.000 description 1
- 238000005452 bending Methods 0.000 description 1
- 238000002485 combustion reaction Methods 0.000 description 1
- 238000010219 correlation analysis Methods 0.000 description 1
- 238000001514 detection method Methods 0.000 description 1
- 230000005489 elastic deformation Effects 0.000 description 1
- 230000003028 elevating effect Effects 0.000 description 1
- 230000002708 enhancing effect Effects 0.000 description 1
- 230000001747 exhibiting effect Effects 0.000 description 1
- 230000006870 function Effects 0.000 description 1
- 238000013178 mathematical model Methods 0.000 description 1
- 238000005457 optimization Methods 0.000 description 1
- 229910052760 oxygen Inorganic materials 0.000 description 1
- 239000001301 oxygen Substances 0.000 description 1
- 238000000513 principal component analysis Methods 0.000 description 1
- 230000000306 recurrent effect Effects 0.000 description 1
- 230000000630 rising effect Effects 0.000 description 1
- 230000035939 shock Effects 0.000 description 1
- 238000004088 simulation Methods 0.000 description 1
- 238000001228 spectrum Methods 0.000 description 1
- 238000012360 testing method Methods 0.000 description 1
- 238000012549 training Methods 0.000 description 1
- 238000012546 transfer Methods 0.000 description 1
Classifications
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F30/00—Computer-aided design [CAD]
- G06F30/10—Geometric CAD
- G06F30/15—Vehicle, aircraft or watercraft design
-
- 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
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06N—COMPUTING ARRANGEMENTS BASED ON SPECIFIC COMPUTATIONAL MODELS
- G06N3/00—Computing arrangements based on biological models
- G06N3/02—Neural networks
- G06N3/04—Architecture, e.g. interconnection topology
- G06N3/045—Combinations of networks
-
- G—PHYSICS
- G06—COMPUTING; CALCULATING OR COUNTING
- G06F—ELECTRIC DIGITAL DATA PROCESSING
- G06F2218/00—Aspects of pattern recognition specially adapted for signal processing
- G06F2218/08—Feature extraction
-
- Y—GENERAL TAGGING OF NEW TECHNOLOGICAL DEVELOPMENTS; GENERAL TAGGING OF CROSS-SECTIONAL TECHNOLOGIES SPANNING OVER SEVERAL SECTIONS OF THE IPC; TECHNICAL SUBJECTS COVERED BY FORMER USPC CROSS-REFERENCE ART COLLECTIONS [XRACs] AND DIGESTS
- Y02—TECHNOLOGIES OR APPLICATIONS FOR MITIGATION OR ADAPTATION AGAINST CLIMATE CHANGE
- Y02T—CLIMATE CHANGE MITIGATION TECHNOLOGIES RELATED TO TRANSPORTATION
- Y02T90/00—Enabling technologies or technologies with a potential or indirect contribution to GHG emissions mitigation
Landscapes
- Engineering & Computer Science (AREA)
- Physics & Mathematics (AREA)
- Theoretical Computer Science (AREA)
- Geometry (AREA)
- General Physics & Mathematics (AREA)
- Evolutionary Computation (AREA)
- General Engineering & Computer Science (AREA)
- Computer Hardware Design (AREA)
- Biophysics (AREA)
- Artificial Intelligence (AREA)
- General Health & Medical Sciences (AREA)
- Molecular Biology (AREA)
- Computing Systems (AREA)
- Computational Linguistics (AREA)
- Biomedical Technology (AREA)
- Mathematical Physics (AREA)
- Software Systems (AREA)
- Data Mining & Analysis (AREA)
- Life Sciences & Earth Sciences (AREA)
- Health & Medical Sciences (AREA)
- Automation & Control Theory (AREA)
- Aviation & Aerospace Engineering (AREA)
- Computational Mathematics (AREA)
- Mathematical Analysis (AREA)
- Mathematical Optimization (AREA)
- Pure & Applied Mathematics (AREA)
- Feedback Control In General (AREA)
Abstract
The invention belongs to automatic control technology fields, and in particular to a kind of method and device of big envelope curve flight Interference Estimation.Aim to solve the problem that the problem of analysis method of prior art rigid aircraft can not be applied to Elastic Vehicles flight dynamics characteristic.The present invention provides a kind of big envelope curve flight interference estimation method, input signal and output signal including obtaining big envelope curve aircraft respectively, Fast Fourier Transform (FFT) signal analysis is carried out to input signal and output signal, the signal that analysis is obtained carries out phase division operation, obtains Bode diagram of the output signal to input signal；Extract the sub-district area image in Bode diagram；The sub-district area image input convolutional neural networks model of Bode diagram is identified, obtains the classification of the sub-district area image of Bode diagram to complete big envelope curve flight Interference Estimation.The method of the present invention can form the mode of oscillation off-line learning mechanism of deep learning mechanism and online robust flexibility recognizes mechanism, be embodied as elastic characteristic Parameter Measuring and calculate.
Description
Technical Field
The invention belongs to the technical field of automatic control, and particularly relates to a method and a device for estimating flight interference of a large envelope curve.
Background
With the continuous expansion of mission range of the aircraft, the flight envelope line of the aircraft is larger and larger. Generally, an aircraft flies in the atmosphere (within 2 ten thousand meters) and in the near space at the edge of the atmosphere (within 2 to 10 ten thousand meters), the flight speed range of the aircraft is between high subsonic speed and high supersonic speed above mach 5, the flight speed change is large, the dynamic characteristic change of the aircraft is also large, and therefore in the flying process of the aircraft, the performance parameters of the aircraft, such as structural elasticity and the like, cannot be ignored, and the design of a flight control law has a great influence on the performance parameters of the aircraft. The control law design method of the aircraft aims to meet the requirement that the aircraft can guarantee the flight quality of the aircraft at different typical working points in an envelope and the aircraft can reach ideal performance indexes in both a time domain and a frequency domain.
As the structural modal frequency and the rigid modal frequency of the aircraft are closer to each other, the influence of the elastic effect on the flight dynamics of the elastic aircraft is more obvious, particularly, the stability control characteristics become more complex and severe, and the analysis method of the rigid aircraft cannot be used for research, so that a flight dynamics model capable of containing multidisciplinary coupling is urgently required to be established for the elastic aircraft, and a targeted control law is designed for the elastic characteristics.
Therefore, how to provide a solution for identifying the characteristic parameters of the large envelope flight control law for the elastic interference is a problem that needs to be solved by those skilled in the art.
Disclosure of Invention
In order to solve the above problems in the prior art, that is, to solve the problem that the analysis method of the rigid aircraft in the prior art cannot be applied to the flight dynamics characteristics of the elastic aircraft, the invention provides a method for estimating flight interference of a large envelope, comprising the following steps:
respectively acquiring an input signal and an output signal of a large-envelope aircraft, and performing Fast Fourier Transform (FFT) signal analysis on the input signal and the output signal to respectively obtain a first FFT signal and a second FFT signal;
performing a division operation on the first FFT signal and the second FFT signal to obtain a bode plot of the output signal versus the input signal;
extracting an image with resonance characteristics in the bode graph as a sub-region image of the bode graph;
and inputting the sub-region image of the bode diagram into a preset convolution neural network model for identification to obtain the category of the sub-region image of the bode diagram so as to complete the estimation of the flight interference of the large envelope curve.
In a preferred technical solution of the above method, the category of the sub-region image of the bode map includes a flat region, a resonance region, and a descent region.
In a preferred embodiment of the above method, the method further comprises: and performing self-adaptive elastic filtering based on the elastic conduction spatial characteristic fusion.
In a preferred embodiment of the foregoing method, the adaptive elastic filtering method includes:
acquiring elastic fluctuation data of the large envelope aircraft through a vibration sensor and recording the spatial position of the vibration sensor;
carrying out speed correction processing on the elastic fluctuation data;
according to the space distribution difference of the sensor array of the large envelope aircraft, solving time domain and frequency domain information of the elastic fluctuation data by adopting a high-dimensional joint spectrum analysis technology;
and establishing a spatial distribution map of a fluctuation curve equation for flight control feedback and stability analysis of the large envelope aircraft according to the elastic fluctuation data, the spatial distribution difference of the sensor array and the time domain and frequency domain information of the elastic fluctuation data.
In a preferred technical scheme of the method, the input signal is an elevator of the large envelope aircraft, and the output signal is a track inclination angle, a height, an attack angle and a pitch angle rate of the large envelope aircraft.
The invention also provides a storage device in which a plurality of programs are stored, said programs being adapted to be loaded by a processor and to carry out the method of estimating a flight disturbance of a large envelope line as described above.
The invention also provides a processing device, which comprises a processor and a storage device; a processor adapted to execute various programs; a storage device adapted to store a plurality of programs; the program is adapted to be loaded by a processor and to perform the method of estimating a flight disturbance of a large envelope line as described above.
Compared with the closest prior art, the invention provides a large envelope flight interference estimation method, which comprises the steps of respectively obtaining an input signal and an output signal of a large envelope aircraft, and carrying out Fast Fourier Transform (FFT) signal analysis on the input signal and the output signal to respectively obtain a first FFT signal and a second FFT signal; performing division operation on the first FFT signal and the second FFT signal to obtain a baud graph of the output signal to the input signal; extracting an image with resonance characteristics in the bode graph as a sub-region image of the bode graph; and inputting the sub-region image of the bode diagram into a preset convolution neural network model for identification to obtain the category of the sub-region image of the bode diagram so as to complete the estimation of the flight interference of the large envelope curve.
The technical scheme at least has the following beneficial effects:
1. the method is characterized in that a closed-loop online robust inversion mechanism of elastic characteristic parameters is designed aiming at elastic interference of a large envelope aircraft, identifiability of closed-loop identification and a robust identification method under data interference are given, and a vibration mode offline learning mechanism and an online robust flexible identification mechanism of a deep learning mechanism are formed;
2. the invention provides a self-adaptive elastic filtering method for the fusion of elastic conduction space characteristics of a large envelope aircraft, and provides an implementation method for the online measurement and calculation of elastic characteristic parameters.
Drawings
FIG. 1 is a graph of the amplitude-frequency characteristics of a control input to a control output according to one embodiment of the present invention;
FIG. 2 is a graph of amplitude-frequency characteristics of a linearized model according to one embodiment of the present invention;
FIG. 3 is a graph of the amplitude-frequency response of the output signal of a large envelope aircraft according to one embodiment of the invention;
FIG. 4 is a graph of amplitude-frequency response of an input signal versus elastic modal output according to an embodiment of the present invention;
FIG. 5 is a diagram illustrating feature extraction based on elastic fluctuation of deep learning according to an embodiment of the present invention;
FIG. 6 is a flowchart illustrating a method for estimating a flight interference of a large envelope curve according to an embodiment of the present invention;
FIG. 7 is a schematic diagram of a coordinate system definition and stress analysis of a node of an elastic vibration sensor and an elastic beam according to an embodiment of the present invention;
fig. 8 is a schematic diagram of adaptive elastic filtering based on elastic conductive spatial characteristic fusion according to an embodiment of the present invention.
Detailed Description
In order to make the objects, technical solutions and advantages of the embodiments of the present invention clearer, the technical solutions in the embodiments of the present invention will be clearly and completely described below with reference to the drawings in the embodiments of the present invention, and it is obvious that the described embodiments are some, but not all, embodiments of the present invention. All other embodiments, which can be obtained by a person skilled in the art without any inventive step based on the embodiments of the present invention, are within the scope of the present invention
Preferred embodiments of the present invention are described below with reference to the accompanying drawings. It should be understood by those skilled in the art that these embodiments are only for explaining the technical principle of the present invention, and are not intended to limit the scope of the present invention.
The invention adopts a calculation method of a robust inversion mechanism of flight elastic characteristic parameters of the large envelope, and analyzes key influence factors of closed-loop cross coupling among uncertainty, rigid body mode, elastic mode and a control system on the dynamics modeling and cross coupling characteristics of uncertainty, elasticity and control of a large envelope aircraft; a closed-loop online robust inversion mechanism of elastic characteristic parameters is designed, identifiability of closed-loop identification and a robust identification method under data interference are provided, a vibration mode offline learning mechanism and an online robust flexible identification mechanism of a deep learning mechanism are formed, and the method is a novel data-driven elastic characteristic parameter inversion mechanism and a calculation implementation method based on elastic characteristic mode extraction.
Meanwhile, an adaptive elastic filtering mechanism based on the fusion of elastic conduction spatial characteristics is formed, the spatial characteristics of the elastic fluctuation conduction mechanism are given, and the time-space distribution of the fluctuation characteristics of different positions of the aircraft is realized; giving a fluctuation characteristic space sampling mode of distributed fluctuation conduction and redundancy characteristics of fluctuation data, constructing a data fusion mechanism, and realizing a calculation method for enhancing and isolating elastic characteristic parameters; a sensing mechanism and a spatial configuration mechanism of fluctuation characteristic measurement are formed, a mathematical expression and a calculation mechanism of elastic fluctuation space estimation facing vibration isolation are given, and an elastic filtering method based on adaptive elastic decoupling and spatial data fusion under the condition of multi-axis elastic characteristic parameter inversion is formed.
Modeling the dynamics of the large envelope aircraft elastomer aircraft, the model can be represented by the following formula:
the model comprises 11 flight states, 5 rigid body states V, gamma, h, α Q respectively representing speed, track inclination angle, altitude, attack angle and pitch angle rate, and 6 elastomer statesRespectively representing the first three elastic modes and their differentials. Wherein, ω isinatural frequency, xi, of elastic modeiM, g, I as damping ratioyyRespectively representing mass, gravitational acceleration and moment of inertia about the Y-axis; l, D, T, M, NiRespectively representing lift force, resistance force, thrust force, pitching moment and generalized elastic force. The model has a complex nonlinear relation with the flight state and the control input, and can be represented by a high-precision fitting model, which is specifically represented by the following formula:
wherein,representing the elastic modal vector, δ ═ δc,δe]TRepresenting the rudder deflection angle vector, δc,δeRespectively showing the rudder deflection angles of the canard and the elevator,represents the dynamic pressure, ρ is the atmospheric density.
The aerodynamic coefficient of a large envelope aircraft can be shown in equation (8):
wherein, CT,φThe thrust coefficient of thrust versus roll angle,are respectively CT,φDerivatives of the angle of attack of order 3, 2, 1 and offset values; cTIs the thrust coefficient of thrust versus angle of attack,are respectively CTFor 3 steps and 2 steps of attack angle,Derivative of order 1 and offset value;
CLcoefficient of lift, angle of attack(elastic mode vector), δ ═ δc,δe]T(rudder deflection angle vector) is related to, deltac,δeRespectively showing the rudder deflection angles of the canard and the elevator,as the lift coefficient of lift to angle of attack,is the lift force to deltaeThe coefficient of lift force of (a) is,is the lift force to deltacThe coefficient of lift force of (a) is,the lift coefficient is zero, and the lift coefficient is zero,the lift coefficient is the lift force to the elastic modal vector η.
CDIs coefficient of resistance, angle of attack(elastic mode vector), δ ═ δc,δe]T(rudder deflection angle vector) is related to, deltac,δeRespectively showing the rudder deflection angles of the canard and the elevator,the drag coefficient of drag to the square of the angle of attack and the angle of attack,is resistance pair deltaeSum of squares deltaeThe coefficient of resistance of (a) is,is resistance pair deltacSum of squares deltacThe coefficient of resistance of (a) is,the resistance coefficient of the material is zero, and the material is,is the resistance coefficient of the resistance to the elastic modal vector eta.
CMIs the coefficient of pitching moment and the angle of attack(elastic mode vector), δ ═ δc,δe]T(rudder deflection angle vector) is related to, deltac,δeRespectively showing the rudder deflection angles of the canard and the elevator,the coefficient of the pitching moment coefficient is the square of the pitching moment to the attack angle and the attack angle,is the pitch moment pair deltaeThe coefficient of the pitching moment of (a),is the pitch moment pair deltacThe coefficient of the pitching moment of (a),the pitch moment coefficient is zero, and the pitch moment coefficient is zero,the pitch moment coefficient is the pitch moment coefficient of the pitch moment coefficient to the elastic modal vector eta.
L,D,T,M,NiRespectively representing lift force, resistance force, thrust force, pitching moment and generalized elastic force.Is the pair of thrust T, lift L, resistance D and pitching moment MThrust coefficient, lift coefficient, drag coefficient, and pitching moment coefficient of (elastic modal vector).Is a generalized elastic force pairThe generalized elastic force coefficient (elastic modal vector).
The canard of a large envelope aircraft is hinged to the elevator to eliminate non-minimum phase characteristics, so the control input to be designed for the aircraft may be u ═ δe,φ]TAnd phi is the engine fuel equivalence ratio. The control output may be y ═ V, h]TI.e. control speed and altitude.
The aircraft mathematical model is a typical multivariable, strong nonlinear and strong coupling model, and due to the fact that the aircraft is high in flying speed, large in flying envelope, complex in mechanism such as scramjet combustion and the like, and lack of sufficient flight test data support, the control model has serious uncertainty and belongs to a typical system with complex structure uncertainty. While uncertainty has a significant impact on model properties. Taking the case of reentry flight, the reentry process is accompanied by rapid velocity and altitude changes, resulting in dynamic pressureThe fast time-varying and uncertain effect of dynamic pressure variations have a significant influence on the flight characteristics.
As shown in FIG. 1, FIG. 1 is a graph of amplitude-frequency characteristics of control input to control output, and a large envelope aircraft is a multidisciplinary intersection product, and complex coupling effects exist among the structure, the propulsion system, the aeroelastic effect, the thermal effect and the control.
From a control point of view, the main coupling effect is represented by the coupling between the engine and the coupling between the rigid-elastic modes. Due to the adoption of the scramjet engine to obtain oxygen materials from air, the overall structure of the large envelope aircraft adopts a high-degree engine-body integrated design.
The front end of the body of the large envelope aircraft is often used as a precompression system of the air inlet flow of an engine, and the rear body is used as an expansion system of the wake flow, so that the posture of the body directly influences the working condition of the engine; in turn, because the engine is often installed at the lower part of the body, the thrust line does not pass through the center of mass, so the thrust generates extra moment, and the shear flow of the rear body also generates extra lift force.
the expression of the thrust T shows that the engine thrust T is obviously influenced by the attack angle alpha no matter the cantilever beam model or the free beam model, so that the high-performance attack angle control is important for the normal work of the scramjet engine, and the coupling between the aerodynamic effect and the working condition of the engine is shown.
At the same time, from pitching momentThe expression shows that the thrust T can generate additional pitching moment with the acting arm zTTherefore, the longitudinal pitching motion is controlled by the fuel equivalence ratio in addition to the elevator, and the cross coupling of the control is shown.
Due to the adoption of the fuselage made of a light material with a large slenderness ratio, the elastic effect of the hypersonic aircraft is more obvious than that of a common aircraft. The elastic deformation of the body will directly change the characteristics of the shock waves of the attachment on the body, on one hand, the aerodynamic characteristics are directly influenced, on the other hand, the air intake condition of the engine is also changed, and therefore the flight characteristics are obviously influenced. The rigid body-elastic coupling effect of the free beam model is taken as an example for analysis:
as known from the pneumatic model, the rigid body-elastic cross coupling in the model is embodied in the expression of force and moment, and the coupling parameter set for defining the elastic mode to the rigid body dynamic is as followsj ═ T, L, D, M; the coupling parameter set for defining rigid body mode to elastic dynamic isi is 1,2, 3. The frequency characteristic of the linearized model is analyzed, and the amplitude-frequency characteristic is shown in fig. 2. The long period mode of slight oscillation can be seen in FIG. 2, corresponding to a frequency of 0.0485rad/s identified at "P". The response of velocity V is not significantly affected by the elastic coupling, but the response of height h exhibits a significant elastic coupling characteristic, denoted as "F1”、“F2”、“F3frequency of place sign and first third order elastic mode η1、η2、η3The natural frequencies of the frequency bands are approximate, and are respectively 21.17 rad/s, 53.92rad/s and 109.1 rad/s.
When the coupling parameters are uncertain, there is a cross-coupling effect, as shown in FIG. 3, FIG. 3 is from δeto the amplitude-frequency response diagram of gamma, h, α q, first consider the parameter set theta1: respectively to theta1The parameters in the method are subjected to uncertainties of-100% (no elastic coupling), 0% (nominal elastic coupling) and + 100% (increased elastic coupling), the corresponding characteristic roots of the linearized model under three conditions are shown in the table, the coupling coefficient of the elastic mode to the rigid mode obviously influences the position of the open loop pole of the model, and the change proportion of the characteristic roots corresponding to the long-period mode and the height mode is larger than that of the short-period mode, so that the conclusion of the coupling strength degree between different rigid modes and the elastic mode can be obtained: the high-frequency wave-splitting mode is a long-period mode, a high-period mode and a short-period mode from strong to weak.
Table 1: theta1The parameters are differentSystem open loop feature root under uncertainty
To further confirm this conclusion, the frequency analysis method was first used to examine the coupling of elastic to rigid modes:
under the condition of + 100% parameter uncertainty, a track inclination angle α (corresponding to a long period mode), a height h (corresponding to a height mode), an attack angle alpha (corresponding to a short period mode) and a pitch angle rate q (corresponding to a short period mode) are respectively selected as single output quantities, an elevator deflection angle is selected as single input quantity, the model is linearized, and then the linearized model is subjected to spectrum characteristic analysis to obtain amplitude-frequency characteristics, as shown in fig. 2, it can be seen from fig. 2 that response characteristics of α and h oscillate at frequencies corresponding to elastic modes, and coupling influence on response characteristics of short period modes (alpha and q) is much smaller.
Conversely, the coupling effect of rigid body modes on elastic modes was examined:
let Θ be2the parameter in the step (a) has-100% uncertainty, and the first three-order elastic modal generalized coordinate value η is respectively selected1、η2、η3For single output, the elevator deflection angle is also selected as single input, and after the model is linearized, the amplitude-frequency characteristic is obtained, as shown in fig. 3, where δ is shown in fig. 3eto eta1、η2、η3Each amplitude-frequency response curve corresponds to two peak points, one of which is at the natural frequency of the elastic mode itself, and the other of which points to a common frequency value (identified by a circle in the figure), namely about 0.0415rad/s, which is very similar to the frequency of the long-period mode.
In conjunction with fig. 3 and 4, it can be concluded that: for an elastic body of a free beam model, the coupling effect between the elastic mode and the long-period mode in the rigid body mode is most obvious, and the strength of cross coupling is clarified between the elastic mode and the height mode and between the elastic mode and the short-period mode, so that the elastic mode can be actively inhibited in a targeted manner during the design of a controller.
By combining the problem to be solved by the invention and the practical application scene, the deep learning is suitable for processing the off-line elastic vibration simulation big data with noise, and the method intensively reflects three big trends of the current machine learning algorithm:
1. reducing model bias with a more complex model;
2. improving the accuracy of statistical estimation by using big data;
3. and solving the large-scale optimization problem by using an extensible gradient descent algorithm.
Deep learning is an end-to-end machine learning system, and can be used for processing a two-dimensional space structure in an image by convolution and processing a time sequence structure in data such as elastic fluctuation and the like by a recurrent neural network. Deep learning can increase the scale of training data under the condition of complex model structure; adding various a priori knowledge about the data structure to the new model structure; end-to-end learning can forego intermediate steps based on artificial rules.
As shown in fig. 5, fig. 5 exemplarily shows a feature extraction diagram of elastic fluctuation based on deep learning. In the embodiment of the invention, elastic fluctuation based on deep learning is adopted for feature extraction, wherein in the aspect of resonance point detection, a scanning window or candidate window method is adopted, the scanning window method shares features between adjacent windows, an image with a large area can be rapidly scanned, the candidate window method can efficiently identify in an image candidate area, and change of an object aspect ratio is more flexibly processed, so that high intersection ratio coverage rate is obtained.
Specifically, as shown in fig. 6, fig. 6 exemplarily shows a flow schematic diagram of a method for estimating flight interference of a large envelope, where the method for inverting elastic feature parameters based on deep learning includes the following steps:
step S11: respectively acquiring an input signal and an output signal of the large-envelope aircraft, and performing Fast Fourier Transform (FFT) signal analysis on the input signal and the output signal to respectively obtain a first FFT signal and a second FFT signal;
acquiring an input signal, the input signal may include: elevating rudder deltaethe output signal may include a track inclination angle γ (corresponding to a long-period mode), a height h (corresponding to a height mode), an attack angle α (corresponding to a short-period mode), and a pitch angle rate q (corresponding to a short-period mode), and then FFT (Fast Fourier transform) signal analysis is performed on the track inclination angle γ (corresponding to a long-period mode), the height h (corresponding to a height mode), the attack angle α (corresponding to a short-period mode), and the pitch angle rate q (corresponding to a short-period mode), respectively, and the rising and falling rudder δ is subjected to FFT (Fast Fourier transform) signal analysiseand performing FFT signal analysis, namely performing FFT signal analysis on the input signal to obtain a result as a first FFT signal, performing FFT signal analysis on the output signal to obtain a result as a second FFT signal, dividing the FFT signals to obtain a Baud diagram of a track inclination angle gamma, a height h, an attack angle α and a pitch angle rate q to a corresponding transfer function of the elevator, and uniformly setting the frequency range of the Baud diagrams to be 0.001-1000 hz. to be used as an input image for deep learning neural network identification.
Step S12: extracting an image with resonance characteristics in the bode graph as a sub-region image of the bode graph;
and (3) sliding the bode graph from the frequency range of 0.001-1000hz in sequence by taking 1.5 octaves as a window on the bode graph, and extracting an image of the bode graph as a sub-region of the bode graph characteristic. This step is mainly used to extract the area of the distortion characteristic of the bode plot with the resonance characteristic.
Step S13: inputting the sub-region image of the bode diagram into a preset convolution neural network model for identification to obtain the category of the sub-region image of the bode diagram so as to complete the estimation of the flight interference of the large envelope curve;
and sending the extracted subregion image into a convolutional neural network for identification. The convolution templates of the convolutional neural network are bode image images of a flat area, a resonance area and a descent area respectively.
Step S14: and (5) classifying the regions.
The sub-regions of the bode plot are classified as flat region Fat, Resonance region Resonance, down region down.
After the inversion of the elastic characteristic parameters is completed based on deep learning, the definition and stress analysis of the elastic vibration sensor node and the elastic beam coordinate system are carried out.
After the elastic dynamic equation is established, a full-modal model equation containing rigid body modes and elastic modes can be established based on Lagrange's equation. However, since elastic aircrafts exhibit more complex dynamics than rigid aircrafts, and the presently disclosed elastic aircraft models have less data, the current research on elastic airframe control is mainly focused on its longitudinal movement, when all lateral conditions are assumed to be 0 in order to avoid adding lateral maneuvers during the flight mission. For the cantilever model, consider the first order elastic mode shapes of the front body (subscript f) and the back body (subscript a) with the equations for longitudinal motion as:
wherein eta isi(i ═ f, a) represents the generalized elastic coordinate value of the first order mode shape,for the coupling coefficient associated with the first-order dominant mode shape, NiFor elastic force in general, omegaiis the elastic natural frequency, ξiIs an elastic damping coefficient, constantOther symbol definitions are the same as rigid body models. The pitch angle rate equation in the model is added with elastic coupling, and the elastic equation is also subjected to pitch momentThereby exhibiting a rigid-elastic dynamic coupling relationship.
The self-adaptive elastic filtering method based on the fusion of the elastic conduction spatial characteristics specifically comprises the following steps:
step S21: sampling by adopting sensors, and recording the corresponding spatial position of each sensor;
due to the strong periodic characteristics of the elastic fluctuation, the response of the elastic fluctuation on the fuselage of the airplane can also show periodicity, such as the spatial bending and twisting positions of the equivalent beam, so that a vibration fluctuation sensor is required to be mounted on the airplane for sampling to acquire data of the sensor, wherein the sensor adopts the vibration fluctuation sensor shown in fig. 7 for sampling.
Step S22: speed correction processing of data;
in practical application, the flying speed is changed, and then the periodic data of the vibration fluctuation can be processed later only after being subjected to speed correction, and the specific method can be that the data acquired by the triaxial accelerometer is divided by the dynamic pressure
Step S23: spectral analysis of the fluctuation signal;
the typical wave motion signal is subjected to spectrum analysis technology (such as spectrum analysis, principal component analysis, autocorrelation analysis and phase time trajectory difference PTPD method), so that a periodic signal of vibration can be extracted; by utilizing the spatial distribution difference of the sensor array (inertial device, gyroscope and accelerometer), and applying a high-dimensional combined spectrum analysis technology (such as time-space evolution correlation analysis and cross spectrum analysis: a linear array BORGMAN method/a polygonal array ESTEVA method), the time domain and frequency domain parameter information of the vibration fluctuation can be obtained.
Step S24: and establishing a spatial distribution diagram of the wave curve equation in the projectile body.
The elastic characteristic parameters estimated in real time, and the spatial distribution parameters and the real-time sensor parameters of the sensor array in fig. 6 can establish the spatial distribution of a wave curve equation in a projectile body, so that the actual attitude parameters, angular rate parameters and vibration isolation calculation of an accelerometer can be performed according to data fusion of multiple sensor points, and further the attitude parameters, angular rate parameters and acceleration parameters with greatly attenuated wave characteristics are obtained, and the parameters and the upper limit of the wave amplitude thereof are used for analyzing flight control feedback and stability.
The steps of a method or algorithm described in connection with the embodiments disclosed herein may be embodied in hardware, a software module executed by a processor, or a combination of the two. A software module may reside in Random Access Memory (RAM), memory, Read Only Memory (ROM), electrically programmable ROM, electrically erasable programmable ROM, registers, hard disk, a removable disk, a CD-ROM, or any other form of storage medium known in the art.
The invention also provides a storage device in which a plurality of programs are stored, said programs being adapted to be loaded by a processor and to perform the method of estimating flight disturbances of a large envelope as described above.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working process and the related descriptions of the storage device according to the embodiment of the present invention may refer to the corresponding process in the foregoing embodiment of the method for estimating flight interference with large envelope, and have the same beneficial effects as the foregoing method for estimating flight interference with large envelope, and are not described herein again.
A processing apparatus comprising a processor, a storage device; a processor adapted to execute various programs; a storage device adapted to store a plurality of programs; the program is adapted to be loaded by a processor and to perform the method of estimating flight disturbances of a large envelope as described above.
It can be clearly understood by those skilled in the art that, for convenience and brevity of description, the specific working process and the related descriptions of the processing apparatus according to the embodiment of the present invention may refer to the corresponding process in the foregoing large envelope flight interference estimation method embodiment, and have the same beneficial effects as the above large envelope flight interference estimation method, and are not described herein again.
Those of skill in the art will appreciate that the method steps of the examples described in connection with the embodiments disclosed herein may be embodied in electronic hardware, computer software, or combinations of both, and that the components and steps of the examples have been described above generally in terms of their functionality in order to clearly illustrate the interchangeability of electronic hardware and software. Whether such functionality is implemented as electronic hardware or software depends upon the particular application and design constraints imposed on the solution. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.
So far, the technical solutions of the present invention have been described in connection with the preferred embodiments shown in the drawings, but it is easily understood by those skilled in the art that the scope of the present invention is obviously not limited to these specific embodiments. Equivalent changes or substitutions of related technical features can be made by those skilled in the art without departing from the principle of the invention, and the technical scheme after the changes or substitutions can fall into the protection scope of the invention.
Claims (7)
1. A method for estimating large envelope flight interference, comprising:
respectively acquiring an input signal and an output signal of a large-envelope aircraft, and performing Fast Fourier Transform (FFT) signal analysis on the input signal and the output signal to respectively obtain a first FFT signal and a second FFT signal;
performing a division operation on the first FFT signal and the second FFT signal to obtain a bode plot of the output signal versus the input signal;
extracting an image with resonance characteristics in the bode graph as a sub-region image of the bode graph;
and inputting the sub-region image of the bode diagram into a preset convolution neural network model for identification to obtain the category of the sub-region image of the bode diagram so as to complete the estimation of the flight interference of the large envelope curve.
2. The method of claim 1, wherein the categories of sub-region images of the bode plot include a flat region, a resonance region, and a dip region.
3. The method of claim 1, further comprising: and performing self-adaptive elastic filtering based on the elastic conduction spatial characteristic fusion.
4. The method of claim 3, wherein the method of adaptive elastic filtering comprises:
acquiring elastic fluctuation data of the large envelope aircraft through a vibration sensor and recording the spatial position of the vibration sensor;
carrying out speed correction processing on the elastic fluctuation data;
according to the space distribution difference of the sensor array of the large envelope aircraft, solving time domain and frequency domain information of the elastic fluctuation data by adopting a high-dimensional joint spectrum analysis technology;
and establishing a spatial distribution map of a fluctuation curve equation for flight control feedback and stability analysis of the large envelope aircraft according to the elastic fluctuation data, the spatial distribution difference of the sensor array and the time domain and frequency domain information of the elastic fluctuation data.
5. The method of claim 1, wherein the input signals are elevators of the large envelope aircraft and the output signals are track pitch, altitude, angle of attack, and pitch rate of the large envelope aircraft.
6. A memory device having stored therein a plurality of programs, characterized in that said programs are adapted to be loaded by a processor and to carry out the method of large envelope flight interference estimation as claimed in any one of claims 1 to 5.
7. A processing apparatus comprising a processor, a storage device; a processor adapted to execute various programs; a storage device adapted to store a plurality of programs; characterized in that the program is adapted to be loaded by a processor and to carry out the method of large envelope flight interference estimation according to any of claims 1-5.
Priority Applications (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810233100.7A CN108595756B (en) | 2018-03-21 | 2018-03-21 | Method and device for estimating flight interference of large envelope |
Applications Claiming Priority (1)
| Application Number | Priority Date | Filing Date | Title |
|---|---|---|---|
| CN201810233100.7A CN108595756B (en) | 2018-03-21 | 2018-03-21 | Method and device for estimating flight interference of large envelope |
Publications (2)
| Publication Number | Publication Date |
|---|---|
| CN108595756A true CN108595756A (en) | 2018-09-28 |
| CN108595756B CN108595756B (en) | 2020-07-28 |
Family
ID=63626978
Family Applications (1)
| Application Number | Title | Priority Date | Filing Date |
|---|---|---|---|
| CN201810233100.7A Active CN108595756B (en) | 2018-03-21 | 2018-03-21 | Method and device for estimating flight interference of large envelope |
Country Status (1)
| Country | Link |
|---|---|
| CN (1) | CN108595756B (en) |
Cited By (6)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109508025A (en) * | 2018-11-23 | 2019-03-22 | 中国科学院数学与系统科学研究院 | A kind of active disturbance rejection attitude control method of Elastic Vehicles |
| CN110244768A (en) * | 2019-07-19 | 2019-09-17 | 哈尔滨工业大学 | Hypersonic aircraft modeling and anti-saturation control method based on switching system |
| CN111583159A (en) * | 2020-05-29 | 2020-08-25 | 北京金山云网络技术有限公司 | Image completion method and device and electronic equipment |
| CN111767668A (en) * | 2020-07-03 | 2020-10-13 | 衢州职业技术学院 | Disc brake squeal noise prediction method and device based on steering knuckle characteristics |
| CN113253616A (en) * | 2021-06-29 | 2021-08-13 | 中国科学院自动化研究所 | Flight control method and device for large envelope of fast time-varying aircraft |
| CN114253308A (en) * | 2020-09-21 | 2022-03-29 | 陕西环保产业研究院有限公司 | Active control method and device for vibration of space frame structure |
Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4002490B2 (en) * | 2001-09-10 | 2007-10-31 | ユーロコプター・ドイッチェランド・ゲゼルシャフト・ミット・ベシュレンクテル・ハフツング | Method for avoiding the rotating rotor blades of a rotorcraft from colliding with the blade wake vortex and apparatus for carrying out the method |
| US20070273863A1 (en) * | 2004-12-17 | 2007-11-29 | The Boeing Company | Ultra-linear signal processing for radar and laser radar |
| CN105022272A (en) * | 2015-07-23 | 2015-11-04 | 北京航空航天大学 | Robustness decoupling control method for elastomer aircraft |
| CN106301623A (en) * | 2016-09-09 | 2017-01-04 | 成都定为电子技术有限公司 | The interference source detection identification method of a kind of spectrum monitoring unmanned plane and device thereof |
| CN106484969A (en) * | 2016-09-23 | 2017-03-08 | 中国运载火箭技术研究院 | A kind of big envelope curve strong maneuvering-vehicle kinetics High Precision Simulation method |
| CN106527128A (en) * | 2016-10-13 | 2017-03-22 | 南京航空航天大学 | Flight control law design new method considering both transient response and robust stability |
| CN106844914A (en) * | 2017-01-09 | 2017-06-13 | 西北工业大学 | A kind of rapid simulation method of re-entry space vehicle wing flutter response |
-
2018
- 2018-03-21 CN CN201810233100.7A patent/CN108595756B/en active Active
Patent Citations (7)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| JP4002490B2 (en) * | 2001-09-10 | 2007-10-31 | ユーロコプター・ドイッチェランド・ゲゼルシャフト・ミット・ベシュレンクテル・ハフツング | Method for avoiding the rotating rotor blades of a rotorcraft from colliding with the blade wake vortex and apparatus for carrying out the method |
| US20070273863A1 (en) * | 2004-12-17 | 2007-11-29 | The Boeing Company | Ultra-linear signal processing for radar and laser radar |
| CN105022272A (en) * | 2015-07-23 | 2015-11-04 | 北京航空航天大学 | Robustness decoupling control method for elastomer aircraft |
| CN106301623A (en) * | 2016-09-09 | 2017-01-04 | 成都定为电子技术有限公司 | The interference source detection identification method of a kind of spectrum monitoring unmanned plane and device thereof |
| CN106484969A (en) * | 2016-09-23 | 2017-03-08 | 中国运载火箭技术研究院 | A kind of big envelope curve strong maneuvering-vehicle kinetics High Precision Simulation method |
| CN106527128A (en) * | 2016-10-13 | 2017-03-22 | 南京航空航天大学 | Flight control law design new method considering both transient response and robust stability |
| CN106844914A (en) * | 2017-01-09 | 2017-06-13 | 西北工业大学 | A kind of rapid simulation method of re-entry space vehicle wing flutter response |
Non-Patent Citations (5)
| Title |
|---|
| ANG LI 等: "LQR and Fuzzy Gain-Scheduling Based Attitude Controller for RLV within Large Operating Envelope", 《2014 IEEE INTERNATIONAL CONFERENCE ON CONTROL SCIENCE AND SYSTEMS ENGINEERING》 * |
| ZHEN LIU 等: "Immersion and Invariance-Based Output Feedback Control of Air-Breathing Hypersonic Vehicles", 《IEEE TRANSACTIONS ON AUTOMATION SCIENCE AND ENGINEERING》 * |
| 卜祥伟 等: "基于非线性干扰观测器的高超声速飞行器滑模反演控制", 《控制理论与应用》 * |
| 王鹏飞 等: "非线性干扰观测器的高超声速飞行器鲁棒反演控制", 《火力与指挥控制》 * |
| 谭湘敏 等: "高超声速飞行器飞行特性分析及其控制研究", 《第二届高超声速科技学术会议》 * |
Cited By (10)
| Publication number | Priority date | Publication date | Assignee | Title |
|---|---|---|---|---|
| CN109508025A (en) * | 2018-11-23 | 2019-03-22 | 中国科学院数学与系统科学研究院 | A kind of active disturbance rejection attitude control method of Elastic Vehicles |
| CN110244768A (en) * | 2019-07-19 | 2019-09-17 | 哈尔滨工业大学 | Hypersonic aircraft modeling and anti-saturation control method based on switching system |
| CN110244768B (en) * | 2019-07-19 | 2021-11-30 | 哈尔滨工业大学 | Hypersonic aircraft modeling and anti-saturation control method based on switching system |
| CN111583159A (en) * | 2020-05-29 | 2020-08-25 | 北京金山云网络技术有限公司 | Image completion method and device and electronic equipment |
| CN111583159B (en) * | 2020-05-29 | 2024-01-05 | 北京金山云网络技术有限公司 | Image complement method and device and electronic equipment |
| CN111767668A (en) * | 2020-07-03 | 2020-10-13 | 衢州职业技术学院 | Disc brake squeal noise prediction method and device based on steering knuckle characteristics |
| CN111767668B (en) * | 2020-07-03 | 2024-03-29 | 衢州职业技术学院 | Steering knuckle characteristic-based disc brake squeal noise prediction method and storage medium |
| CN114253308A (en) * | 2020-09-21 | 2022-03-29 | 陕西环保产业研究院有限公司 | Active control method and device for vibration of space frame structure |
| CN114253308B (en) * | 2020-09-21 | 2022-08-30 | 陕西环保产业研究院有限公司 | Active control method and equipment for vibration of space frame structure |
| CN113253616A (en) * | 2021-06-29 | 2021-08-13 | 中国科学院自动化研究所 | Flight control method and device for large envelope of fast time-varying aircraft |
Also Published As
| Publication number | Publication date |
|---|---|
| CN108595756B (en) | 2020-07-28 |
Similar Documents
| Publication | Publication Date | Title |
|---|---|---|
| CN108595756B (en) | Method and device for estimating flight interference of large envelope | |
| CN108196532B (en) | Fault detection and separation method for longitudinal flight control system of unmanned aerial vehicle based on nonlinear adaptive observer | |
| Morelli et al. | Real-time dynamic modeling: data information requirements and flight-test results | |
| Sankaralingam et al. | A comprehensive survey on the methods of angle of attack measurement and estimation in UAVs | |
| Mohamed et al. | Bioinspired wing-surface pressure sensing for attitude control of micro air vehicles | |
| Pfifer et al. | System Identification of a small flexible aircraft-invited | |
| KR101842966B1 (en) | A measurement method of dynamic stability derivatives of a flight vehicle by utilizing magnetic force | |
| Ahsan et al. | System identification of an airship using trust region reflective least squares algorithm | |
| Mohamed et al. | Influence of turbulence on MAV roll perturbations | |
| Licitra et al. | Optimal input design for autonomous aircraft | |
| KR20220096658A (en) | Method and apparatus for improving the accuracy of 6-dof flight simulation using flight test data | |
| CN106774385B (en) | A kind of dirigible spot hover control method using adaptive variable structure | |
| Licitra et al. | Aerodynamic model identification of an autonomous aircraft for airborne wind energy | |
| Roy et al. | Adaptive backstepping controller for altitude control of a small scale helicopter by considering the ground effect compensation | |
| Morelli et al. | Aircraft dynamic modeling in turbulence | |
| Koehl et al. | Wind-disturbance and aerodynamic parameter estimation of an experimental launched micro air vehicle using an EKF-like observer | |
| KEHOE | Aircraft flight flutter testing at the NASA Ames-Dryden flight research facility | |
| Roy et al. | Robust altitude control for a small helicopter by considering the ground effect compensation | |
| Millidere et al. | Kalman based neural network model analysis with resampling methods for longitudinal aerodynamic coefficient estimation | |
| CN108446461B (en) | Aircraft flutter analysis grid model modeling method | |
| Kotikalpudi et al. | Real-time shape estimation for flexible unmanned air vehicle via Kalman filtering | |
| Rijks et al. | Studying the effect of the tail on the dynamics of a flapping-wing MAV using free-flight data | |
| Szymanski et al. | Experimental Investigation of the Static and Dynamic Stability Derivatives Using a Novel 3-DoF Mechanism | |
| Montano et al. | Automatic wind identification for UAS: a case study | |
| Saderla et al. | Lateral directional parameter estimation of a miniature unmanned aerial vehicle using maximum likelihood and Neural Gauss Newton methods |
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 |