Method for constructing an incremental fusion high angle of attack aerodynamic model based on machine deep learning, electronic equipment and storage medium

Abstract

The method, electronic device and storage medium for constructing an incremental fusion high angle of attack aerodynamic model based on machine deep learning belong to the field of aerodynamics and flight mechanics. In order to improve the prediction accuracy of high angle of attack aerodynamics, the present invention includes obtaining static aerodynamic test data and dynamic derivative test data through wind tunnel tests, obtaining unsteady test data through large oscillation tests or unsteady motion test wind tunnel tests, calculating the static component and dynamic derivative component under the unsteady test motion history, and using the normalized unsteady test data to deduct the static component and dynamic derivative component under the unsteady test motion history to obtain the unsteady aerodynamic increment; using a machine deep learning algorithm to model the unsteady aerodynamic increment, obtain an unsteady aerodynamic coefficient increment model, and then construct a high angle of attack aerodynamic model. The present invention can adapt to a variety of application scenarios from traditional theoretical analysis to simulation analysis.

Description

Technical Field

[0001] The present invention belongs to the field of aerodynamics and flight mechanics, and specifically relates to a method for constructing an incremental fusion high-angle-of-attack aerodynamic model based on machine deep learning, an electronic device, and a storage medium. Background Art

[0002] The aerodynamic force of an aircraft at high angle of attack has strong nonlinear unsteady characteristics. The study of high angle of attack aerodynamic force prediction methods is a traditional problem in the prediction of high angle of attack aerodynamic force of advanced layout aircraft. The existing nonlinear unsteady aerodynamic force prediction models can be roughly divided into two categories. One is a mathematical model, such as aerodynamic derivative model, step response model, state space model and differential equation model. The other is an artificial intelligence model, such as neural network model, fuzzy logic model and support vector machine model. In recent years, machine deep learning algorithms have become the focus of research on high angle of attack aerodynamic modeling. However, in the existing machine deep learning modeling process, only large oscillation data is used for model training. The established model usually has the problem of inaccurate description of static and dynamic derivative characteristics. This is because the static and dynamic derivative characteristics in the model are also learned from large oscillation test data. The most direct impact of this problem is that the established model will cause serious deviations in traditional flight mechanics analysis and simulation, such as static stability deviation, dynamic derivative damping characteristic deviation, etc., making it difficult to carry out engineering applications of the model.

[0003] Therefore, it is necessary to develop an incremental fusion high angle of attack aerodynamic modeling method based on machine deep learning to improve the accuracy of high angle of attack aerodynamic prediction. Summary of the invention

[0004] The problem to be solved by the present invention is to improve the accuracy of high angle of attack aerodynamic prediction, and propose a method for constructing an incremental fusion high angle of attack aerodynamic model based on machine deep learning, an electronic device and a storage medium.

[0005] To achieve the above object, the present invention is implemented through the following technical solutions:

[0006] A method for constructing an incremental fusion high angle of attack aerodynamic model based on machine deep learning, comprising the following steps:

[0007] S1. Obtain static aerodynamic test data through wind tunnel tests, including change data of angle of attack, sideslip angle, and control surface deflection;

[0008] S2. Obtain dynamic derivative test data through wind tunnel tests, including the change data of pitch angular velocity, roll angular velocity, yaw angular velocity and wash flow time difference angular velocity;

[0009] S3. Obtain unsteady test data by using large-scale oscillation test or unsteady motion test wind tunnel test, including aerodynamic data of different oscillation frequency and amplitude combinations within the target aircraft motion frequency and attitude angle envelope;

[0010] S4. Perform data analysis and preprocessing on the static aerodynamic test data obtained in step S1, the dynamic derivative test data obtained in step S2, and the unsteady test data obtained in step S3, and correct wild values ​​and jump points in the data;

[0011] S5. Calculate the static component C under the unsteady test motion history using the pre-processed static aerodynamic test data and dynamic derivative test data obtained in step S4 i,st (α, β, δ) and dynamic derivative components Among them, α is the angle of attack, β is the sideslip angle, and δ is the deflection of the control surface. is the dimensionless pitch angular velocity, is the dimensionless roll angular velocity, is the dimensionless yaw angular velocity, is the dimensionless angle of attack velocity, is the dimensionless sideslip angle rate;

[0012] S6. Using the normalized unsteady test data obtained in step S4, the static component and the dynamic derivative component under the unsteady test motion history obtained in step S5 are deducted to obtain the unsteady aerodynamic force increment;

[0013] S7. Use a machine deep learning algorithm to model the unsteady aerodynamic force increment obtained in step S6, obtain an unsteady aerodynamic force coefficient increment model, and then construct a high angle of attack aerodynamic force model.

[0014] Furthermore, the calculation formula of the unsteady aerodynamic force increment in step S6 is:

[0015]

[0016] in, is the unsteady aerodynamic force increment, C i The data are unsteady test data.

[0017] Furthermore, the machine deep learning algorithm used in step S7 is a long short-term memory recurrent neural network structure, and the input parameters are set to angle of attack, sideslip angle, pitch angular velocity, yaw angular velocity, roll angular velocity and wash flow time difference angular velocity. The deep learning algorithm loss function selects the mean absolute error function MAE, and the calculation formula is:

[0018]

[0019] Among them, m is the total number of data points, is the predicted value of the unsteady increment model, It is the incremental test value of the unsteady aerodynamic test;

[0020] The activation function selects the hyperbolic tangent function tanh(x), and the expression is:

[0021] tanh(x)=(e x -e (-x) ) / (e x +e (-x) ).

[0022] Furthermore, the expression of the unsteady aerodynamic coefficient increment model obtained in step S7 is:

[0023]

[0024] in, is the unsteady aerodynamic coefficient increment model;

[0025] Then the high angle of attack aerodynamic model is obtained, and the formula is constructed as follows:

[0026]

[0027] Among them, C i ′ is the aerodynamic model at high angle of attack.

[0028] Furthermore, large oscillations, maneuvering histories and virtual flight test data were used to verify the prediction results of the high angle of attack aerodynamic model.

[0029] An electronic device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of a method for constructing an incremental fusion high angle of attack aerodynamic model based on machine deep learning are implemented.

[0030] A computer-readable storage medium stores a computer program, which, when executed by a processor, implements a method for constructing an incremental fusion high-angle-of-attack aerodynamic model based on machine deep learning.

[0031] Beneficial effects of the present invention:

[0032] The method for constructing an incremental fusion high-angle-of-attack aerodynamic model based on machine deep learning described in the present invention establishes an incremental fusion high-angle-of-attack aerodynamic model based on machine deep learning. Compared with the existing artificial intelligence high-angle-of-attack aerodynamic model, accurate basic aerodynamic characteristics are given by static and dynamic derivative data, and the incremental characteristics of the aerodynamic force generated by the unsteady state are given by the unsteady aerodynamic model based on deep learning, thereby realizing a comprehensive aerodynamic characteristic description capability from steady to unsteady, solving the problem of inaccurate prediction of static and dynamic derivative data by the existing artificial intelligence high-angle-of-attack aerodynamic model, and can adapt to a variety of application scenarios from traditional theoretical analysis to simulation analysis. BRIEF DESCRIPTION OF THE DRAWINGS

[0033] Figure 1 A flow chart of a method for constructing an incremental fusion high angle of attack aerodynamic model based on machine deep learning according to the present invention;

[0034] Figure 2 This is a comparison curve of the application effect of the model of the present invention. DETAILED DESCRIPTION

[0035] In order to make the purpose, technical solution and advantages of the present invention more clear, the present invention is further described in detail below in conjunction with the accompanying drawings and specific embodiments. It should be understood that the specific embodiments described herein are only used to explain the present invention and are not used to limit the present invention, that is, the specific embodiments described are only part of the embodiments of the present invention, rather than all of the specific embodiments. The components of the specific embodiments of the present invention described and shown in the drawings herein can be arranged and designed in various different configurations, and the present invention can also have other embodiments.

[0036] Therefore, the following detailed description of the specific embodiments of the present invention provided in the accompanying drawings is not intended to limit the scope of the invention claimed for protection, but merely represents the selected specific embodiments of the present invention. Based on the specific embodiments of the present invention, all other specific embodiments obtained by those skilled in the art without making creative work are within the scope of protection of the present invention.

[0037] In order to further understand the content, features and effects of the present invention, the following specific implementation methods are given as examples, and the attached Figure 1 and attached Figure 2 The detailed instructions are as follows:

[0038] Embodiment 1:

[0039] A method for constructing an incremental fusion high angle of attack aerodynamic model based on machine deep learning, comprising the following steps:

[0040] S1. Obtain static aerodynamic test data through wind tunnel tests, including change data of angle of attack, sideslip angle, and control surface deflection;

[0041] S2. Obtain dynamic derivative test data through wind tunnel tests, including the change data of pitch angular velocity, roll angular velocity, yaw angular velocity and wash flow time difference angular velocity;

[0042] S3. Obtain unsteady test data by using large-scale oscillation test or unsteady motion test wind tunnel test, including aerodynamic data of different oscillation frequency and amplitude combinations within the target aircraft motion frequency and attitude angle envelope;

[0043] S4. Perform data analysis and preprocessing on the static aerodynamic test data obtained in step S1, the dynamic derivative test data obtained in step S2, and the unsteady test data obtained in step S3, and correct wild values ​​and jump points in the data;

[0044] S5. Calculate the static component C under the unsteady test motion history using the pre-processed static aerodynamic test data and dynamic derivative test data obtained in step S4 i,st (α, β, δ) and dynamic derivative components Among them, α is the angle of attack, β is the sideslip angle, and δ is the deflection of the control surface. is the dimensionless pitch angular velocity, is the dimensionless roll angular velocity, is the dimensionless yaw angular velocity, is the dimensionless angle of attack velocity, is the dimensionless sideslip angle rate;

[0045] S6. Using the normalized unsteady test data obtained in step S4, the static component and the dynamic derivative component under the unsteady test motion history obtained in step S5 are deducted to obtain the unsteady aerodynamic force increment;

[0046] Furthermore, the calculation formula of the unsteady aerodynamic force increment in step S6 is:

[0047]

[0048] in, is the unsteady aerodynamic force increment, C i The data are unsteady test data.

[0049] S7. Use a machine deep learning algorithm to model the unsteady aerodynamic force increment obtained in step S6, obtain an unsteady aerodynamic force coefficient increment model, and then construct a high angle of attack aerodynamic force model.

[0050] Furthermore, the machine deep learning algorithm used in step S7 is a long short-term memory recurrent neural network structure, and the input parameters are set to angle of attack, sideslip angle, pitch angular velocity, yaw angular velocity, roll angular velocity and wash flow time difference angular velocity. The deep learning algorithm loss function selects the mean absolute error function MAE, and the calculation formula is:

[0051]

[0052] Among them, m is the total number of data points, is the predicted value of the unsteady increment model, It is the incremental test value of the unsteady aerodynamic test;

[0053] The activation function selects the hyperbolic tangent function tanh(x), and the expression is:

[0054] tanh(x)=(e x -e (-x) ) / (e x +e (-x) ).

[0055] Furthermore, the expression of the unsteady aerodynamic coefficient increment model obtained in step S7 is:

[0056]

[0057] in, is the unsteady aerodynamic coefficient increment model;

[0058] Then the high angle of attack aerodynamic model is obtained, and the formula is constructed as follows:

[0059]

[0060] Among them, C i ′ is the aerodynamic model at high angle of attack.

[0061] Furthermore, large oscillations, maneuvering histories and virtual flight test data were used to verify the prediction results of the high angle of attack aerodynamic model.

[0062] Furthermore, the model output is one of the six aerodynamic forces / moments, that is, each aerodynamic force / moment model is independent.

[0063] Embodiment 2:

[0064] An electronic device comprises a memory and a processor, wherein the memory stores a computer program, and when the processor executes the computer program, the steps of a method for constructing an incremental fusion high angle of attack aerodynamic model based on machine deep learning as described in Example 1 are implemented.

[0065] The computer device of the present invention may be a device including a processor and a memory, such as a single chip microcomputer including a central processing unit. Furthermore, the processor is used to implement the steps of the method for constructing an incremental fusion high angle of attack aerodynamic model based on machine deep learning when executing the computer program stored in the memory.

[0066] The processor may be a central processing unit (CPU), or other general-purpose processors, digital signal processors (DSP), application-specific integrated circuits (ASIC), off-the-shelf programmable gate arrays (FPGAs), or a processor.

[0067] Field-Programmable Gate Array (FPGA) or other programmable logic devices, discrete gate or transistor logic devices, discrete hardware components, etc. The general processor can be a microprocessor or the processor can also be any conventional processor, etc.

[0068] The memory may mainly include a program storage area and a data storage area, wherein the program storage area may store an operating system, an application required for at least one function (such as a sound playback function, an image playback function, etc.), etc.; the data storage area may store data created according to the use of the mobile phone (such as audio data, a phone book, etc.), etc. In addition, the memory may include a high-speed random access memory, and may also include a non-volatile memory, such as a hard disk, a memory, a plug-in hard disk, a smart memory card (SmartMedia Card, SMC), a secure digital (Secure Digital, SD) card, a flash card (Flash Card), at least one disk storage device, a flash memory device, or other volatile solid-state storage devices.

[0069] Embodiment 3:

[0070] A computer-readable storage medium having a computer program stored thereon, wherein when the computer program is executed by a processor, the method for constructing an incremental fusion high angle of attack aerodynamic model based on machine deep learning as described in Example 1 is implemented.

[0071] The computer-readable storage medium of the present invention can be any form of storage medium that can be read by the processor of a computer device, including but not limited to non-volatile memory, volatile memory, ferroelectric memory, etc. The computer-readable storage medium stores a computer program. When the processor of the computer device reads and executes the computer program stored in the memory, the steps of the above-mentioned method for constructing an incremental fusion high-angle-of-attack aerodynamic model based on machine deep learning can be implemented.

[0072] The computer program includes computer program code, which may be in source code form, object code form, executable file or some intermediate form, etc. The computer readable medium may include: any entity or device capable of carrying the computer program code, recording medium, USB flash drive, mobile hard disk, magnetic disk, optical disk, computer memory, read-only memory (ROM), random access memory (RAM), electric carrier signal, telecommunication signal and software distribution medium, etc. It should be noted that the content contained in the computer readable medium may be appropriately increased or decreased according to the requirements of legislation and patent practice in the jurisdiction. For example, in some jurisdictions, according to legislation and patent practice, computer readable media do not include electric carrier signals and telecommunication signals.

[0073] It should be noted that relational terms such as "first" and "second" are used only to distinguish one entity or operation from another entity or operation, and do not necessarily require or imply any such actual relationship or order between these entities or operations. Moreover, the terms "include", "comprise" or any other variants thereof are intended to cover non-exclusive inclusion, so that a process, method, article or device including a series of elements includes not only those elements, but also other elements not explicitly listed, or also includes elements inherent to such process, method, article or device. In the absence of further restrictions, the elements defined by the sentence "comprise a ..." do not exclude the existence of other identical elements in the process, method, article or device including the elements.

[0074] Although the present application has been described above with reference to specific embodiments, various modifications may be made thereto and parts thereof may be replaced with equivalents without departing from the scope of the present application. In particular, as long as there is no structural conflict, the various features in the specific embodiments disclosed in the present application may be used in combination with each other in any manner, and the fact that these combinations are not exhaustively described in this specification is only for the sake of omitting space and saving resources. Therefore, the present application is not limited to the specific embodiments disclosed herein, but includes all technical solutions falling within the scope of the claims.

Claims

1. A method for constructing an incremental fusion high angle of attack aerodynamic model based on machine deep learning, characterized in that: The steps include: S1. Obtain static aerodynamic test data through wind tunnel tests, including change data of angle of attack, sideslip angle, and control surface deflection; S2. Obtain dynamic derivative test data through wind tunnel tests, including the change data of pitch angular velocity, roll angular velocity, yaw angular velocity and wash flow time difference angular velocity; S3. Obtain unsteady test data by using large-scale oscillation test or unsteady motion test wind tunnel test, including aerodynamic data of different oscillation frequency and amplitude combinations within the target aircraft motion frequency and attitude angle envelope; S4. Perform data analysis and preprocessing on the static aerodynamic test data obtained in step S1, the dynamic derivative test data obtained in step S2, and the unsteady test data obtained in step S3, and correct wild values ​​and jump points in the data; S5. Calculate the static component C under the unsteady test motion history using the pre-processed static aerodynamic test data and dynamic derivative test data obtained in step S4 i,st (α, β, δ) and dynamic derivative components Among them, α is the angle of attack, β is the sideslip angle, and δ is the deflection of the control surface. is the dimensionless pitch angular velocity, is the dimensionless roll angular velocity, is the dimensionless yaw angular velocity, is the dimensionless angle of attack velocity, is the dimensionless sideslip angle rate; S6. Using the normalized unsteady test data obtained in step S4, the static component and the dynamic derivative component under the unsteady test motion history obtained in step S5 are deducted to obtain the unsteady aerodynamic force increment; The calculation formula of the unsteady aerodynamic force increment in step S6 is: in, is the unsteady aerodynamic force increment, C i is the unsteady test data; S7. Modeling the unsteady aerodynamic force increment obtained in step S6 using a machine deep learning algorithm to obtain an unsteady aerodynamic force coefficient increment model, and then constructing a high angle of attack aerodynamic force model; The machine deep learning algorithm used in step S7 is a long short-term memory recurrent neural network structure, and the input parameters are set to angle of attack, sideslip angle, pitch angular velocity, yaw angular velocity, roll angular velocity and wash flow time difference angular velocity. The deep learning algorithm loss function selects the mean absolute error function MAE, and the calculation formula is: Among them, m is the total number of data points, is the predicted value of the unsteady increment model, It is the incremental test value of the unsteady aerodynamic test; The activation function selects the hyperbolic tangent function tanh(x), and the expression is: tanh(x)=(and x -And (-x) ) / (And x +e (-x) ); The expression of the unsteady aerodynamic coefficient increment model obtained in step S7 is: in, is the unsteady aerodynamic coefficient increment model; Then the high angle of attack aerodynamic model is obtained, and the formula is constructed as follows: Among them, C i ′ is the aerodynamic model at high angle of attack.

2. The method for constructing an incremental fusion high angle of attack aerodynamic model based on machine deep learning according to claim 1, characterized in that: Large oscillation, maneuver history and virtual flight test data are used to verify the prediction results of the high angle of attack aerodynamic model.

3. An electronic device, characterized in that: It comprises a memory and a processor, the memory stores a computer program, and the processor implements the steps of a method for constructing an incremental fusion high angle of attack aerodynamic model based on machine deep learning as described in any one of claims 1-2 when executing the computer program.

4. A computer-readable storage medium having a computer program stored thereon, characterized in that: When the computer program is executed by a processor, it implements the method for constructing an incremental fusion high angle of attack aerodynamic model based on machine deep learning as described in any one of claims 1-2.

Images