Wind tunnel experiment multi-source hybrid noise effective elimination method based on deep learning algorithm

Abstract

The invention belongs to the technical field of aircraft model support system buffeting detection, and discloses a wind tunnel experiment multi-source hybrid noise effective elimination method based on a deep learning algorithm, and the method comprises the steps: building a force balance buffeting estimation model according to an Euler-Bernoulli theory; solving a vibration kinetic equation by using a variable separation method, and performing modal analysis; and finally, according to the solution of the obtained kinetic equation and a modal analysis result, vibration decoupling and noise elimination are carried out based on a deep learning algorithm. Compared with a finite element analysis method and an analytical calculation method, the method has the advantages that the method is based on a deep learning algorithm, solving operation of a vibration kinetic equation is carried out through a modal analysis method, and compared with the two methods, the method has the advantages of being high in calculation speed, few in consumed calculation resources, high in calculation precision, relatively simple in solving process and the like. The wind tunnel experiment multi-source mixed noise effective elimination method based on the deep learning algorithm is a simple and practical method for a common wind tunnel test scene.

Description

Technical Field

[0001] The present invention belongs to the technical field of buffeting detection of aircraft model support systems, and in particular relates to an effective method for eliminating multi-source mixed noise in wind tunnel experiments based on a deep learning algorithm. Background Art

[0002] For national strategic equipment such as supersonic fighters, hypersonic aircraft, rockets and other high-performance aircraft. Such aircraft often fly at high speed and high maneuverability in a strong flow field environment, and the aerodynamic loads are complex and changeable, and their stability and service safety are seriously threatened. In order to ensure the stability of the aircraft and the reliability of various technical indicators, it is necessary to accurately measure the multi-dimensional aerodynamic force in a high-speed wind tunnel, and then accurately obtain the aerodynamic data of the aircraft. Taking the dynamic derivative test of a certain hypersonic aircraft in a pulse combustion wind tunnel as an example, it is necessary to complete the aerodynamic measurement under continuous multi-directional excitation to simulate the high maneuverability and disturbance conditions of the aircraft. The test wind speed reaches 5~11 times the speed of sound (Mach / Ma), the test time is only 200ms, the measurement error is less than 2%, and the allowable installation space for the internal excitation device and the measuring balance of the large-scale wind tunnel model is only 150mm×50mm×50mm. In particular, the turbulent flow field is complex under large angle of attack conditions, and the model support system produces strong vibration interference, resulting in an overload rate of the measurement system of up to 300%. Among them, under extreme test conditions such as high angle of attack, the effects of various complex flow fields such as wing airflow separation and pulsating airflow cause the wind tunnel model-support system to produce forced strong vibrations. In addition, the test space is limited, and the test has high requirements for measurement dimensions and frequency response. As a result, the aerodynamic measurement data in actual wind tunnel tests have multi-source noise mixing and low aerodynamic data accuracy. Aerodynamic measurement data are often accompanied by multi-source mixed noises such as additional jitter noise, vibration suppression residual vibration noise, and drive oscillation noise, and the signal-noise frequency overlap timing varies. In order to solve the above problems, it is urgent to propose a new method for effectively eliminating multi-source mixed noise in wind tunnel experiments.

[0003] At present, domestic and foreign scholars' research on the removal of multi-source mixed noise in wind tunnel experiments mainly focuses on finite element analysis and analytical calculation methods. Among them, the advantages of finite element analysis are high calculation accuracy, accurate vibration type analysis, and simple operation, but there are problems such as long calculation time, which is not conducive to the real-time monitoring of the experimental system and the application of noise removal; the advantages of analytical calculation method are short calculation time, less consumption of computing resources, and convenient implementation of noise removal algorithm in the experimental process, but there are defects such as poor calculation accuracy, narrow applicability, and inaccurate vibration type analysis.

[0004] Li Yong et al.'s patent "A finite element simulation method for the external field distribution of transformer core vibration noise", patent number CN202110819734.2, introduces a finite element simulation method for the external field distribution of transformer core vibration noise. This method is based on the Helmholtz wave equation and the Fourier transform method of the electromagnetic field. It uses the COMSOL finite element analysis software to perform a multi-field coupling finite element analysis modeling calculation of the magnetic field and solid field of the oil-immersed transformer core vibration noise, and obtains the vibration simulation image of the object. This method provides a simple and effective vibration simulation modeling method, but its limitation is that it only analyzes the vibration of the target object without decoupling the vibration mode, so it cannot meet the engineering requirements of analyzing external mixed noise and eliminating it.

[0005] Ren Zhuoyu in "Based on Wilson- In the article "Research on Load Identification by Using the Method and Trend Term Removal Technology", based on Wilson- This method aims to eliminate the high-level Gaussian noise generated under dynamic loads. It introduces a five-point cubic smoothing method, mathematically fits the load spectrum and the noise term, and decouples the loading response according to the kinematic equation, thereby realizing the identification and elimination of the noise term. This method has the advantages of accurate modeling for specific dynamic loads, high mathematical calculation accuracy, and shorter calculation time than the finite element method. However, it has the limitations of a narrow applicable load spectrum and cannot adapt to complex load conditions in wind tunnel experiments.

[0006] Based on the problems existing in the above technologies, it is necessary to propose an effective method for eliminating multi-source mixed noise in wind tunnel experiments based on deep learning algorithms, shortening the calculation time of the traditional finite element method, and making use of the high adaptability and high precision of deep learning algorithms in the field of vibration noise detection and analysis to perform appropriate physical image modeling, so as to solve the defects of poor calculation accuracy, narrow applicability and inaccurate vibration mode analysis of analytical calculation methods. Summary of the invention

[0007] The purpose of the present invention is to provide a method for effectively eliminating multi-source mixed noise in wind tunnel experiments based on a deep learning algorithm, so as to overcome the shortcomings of the prior art and solve the various problems existing in the existing methods.

[0008] The technical solution of the present invention:

[0009] An effective method for eliminating multi-source mixed noise in wind tunnel experiments based on deep learning algorithms. First, a force balance buffeting estimation model is established based on the Euler-Bernoulli theory. Second, the vibration dynamics equation is solved using the separation of variables method, and modal analysis is performed. Finally, vibration decoupling and noise elimination are performed based on the deep learning algorithm according to the solution of the obtained dynamics equation and the modal analysis results.

[0010] The specific steps are as follows:

[0011] The first step is to establish a force balance buffeting estimation model;

[0012] Consider an experimental system consisting of a force balance and an aircraft model. The two ends of the force balance support rod are fixedly connected to the force balance and the aircraft model respectively. The center of the circle corresponding to the tail cross section of the force balance support rod fixed to the force balance is taken as the origin of the coordinate system, and its axis is taken as Axis, the direction points from the origin to the opposite direction of the force balance support rod;

[0013] According to the theory of elastic mechanics, the force balance support rod is equivalent to a variable cross-section elastic beam, and the differential equation of the forced vibration motion of the variable cross-section elastic beam is:

[0014]

[0015] Where: is the bending moment function, is the time term, is the density of the force balance rod, is the cross-sectional area function of the force balance, is the exciting force function, is the deflection function;

[0016] According to the Euler-Bernoulli theory, the differential equation of the deflection vibration response of the variable cross-section beam is obtained, which is the buffeting estimation model of the force measuring balance:

[0017]

[0018] Where: is the Young's modulus of the force balance material, is the moment of inertia of the cross section of the force balance;

[0019] At this point, the buffeting estimation model for the force balance has been established;

[0020] The second step is to solve the vibration dynamics equation and perform modal analysis;

[0021] According to the separation of variables method for solving partial differential equations, let

[0022]

[0023] Where: is the position-dependent function term in the deflection function, called the vibration mode function; is the time-dependent function term in the deflection function;

[0024] say for The order natural vibration mode function satisfies the characteristic equation:

[0025]

[0026] Using the orthogonality of the vibration mode function, we get:

[0027]

[0028] Where: for The generalized force corresponding to the order vibration mode is, is the length of the variable cross-section elastic beam;

[0029] Using Duhamel points, we get:

[0030]

[0031]

[0032] Where: for The modal amplitude corresponding to the order natural vibration mode is, and They are The two constants under the order natural vibration mode are: for The angular frequency corresponding to the order natural vibration mode is, is the integral parameter, is a constant parameter;

[0033] The natural vibration function called The modal shape function, each vibration mode corresponds to a vibration shape;

[0034] At this point, the solution of the vibration dynamics equations and the modal analysis are completed;

[0035] Step 3: Vibration decoupling and noise removal based on deep learning algorithm

[0036] From formula (6), we know that the solution of the vibration mode generalized coordinate function is The order vibration mode and the corresponding integral constant are expressed. Therefore, based on the deep learning algorithm, according to the fitting function data of the acceleration measurement point in the experiment, the modal order to be determined is set to determine the integral constant. The relationship between the acceleration of the measurement point and the modal amplitude function is:

[0037]

[0038] Where: , Respectively Point and The acceleration at the point, represents the second-order derivative function of the fitted modal amplitude, and due to It is the fitting function of the experimental acceleration measurement point data, and is marked as an estimated quantity;

[0039] A GRU deep learning model is established, where the number of hidden layers of GRU is 2, the number of neurons is [3, 6], the number of hidden layers of the fully connected layer is 3, the number of neurons is [12, 6, 3], and the number of neurons in the output layer is 1; its estimated maximum error is 6.82×10-3g, the average absolute error is 9.20×10-4g, and the root mean square error is 1.2×10-3g, where g represents the gravitational acceleration; the vibration acceleration at the force balance is estimated to achieve vibration decoupling, and the numerical solutions of each mode of equation (6) are determined by fitting the measurement point data;

[0040] Due to the inertia and gravity of the aircraft model, a support reaction force will be generated on the force balance support rod, thereby generating a shear force in the lift direction on the force balance; since the center of mass of the aircraft model does not coincide with the measurement center, a support moment will be generated under the influence of gravity; and a pitch inertia moment will be generated during the oscillation of the aircraft model. Based on Newton's theorem and the parallel axis theorem, its function form is listed as follows:

[0041]

[0042]

[0043] Where: , is the chattering noise value of force and torque, is the gravity acting on the aircraft model, is the angle between the aircraft model and the horizontal plane at the support point of the strut, is the inertial acceleration of the aircraft model, is the angular acceleration of the aircraft model, L1 is the distance from the center of gravity of the aircraft model to the origin of the coordinate system, L2 is the distance from the center of gravity of the aircraft model to the sensor, is the mass of the aircraft model;

[0044] Combining equations (6), (7), (8), and (9) with the fitting results of the measurement point data obtained by the GRU deep learning model, the numerical solutions of the two vibration noises are obtained, which are then decoupled from the experimental data to obtain more accurate experimental results.

[0045] At this point, vibration decoupling and noise removal based on deep learning algorithm are completed.

[0046] The beneficial effect of the present invention is that it proposes a method for effectively eliminating multi-source mixed noise in wind tunnel experiments based on a deep learning algorithm. The difference from the finite element analysis method and the analytical calculation method is that this method is based on a deep learning algorithm and solves the vibration dynamics equations through the modal analysis method. Compared with the above two methods, it has the advantages of fast calculation speed, less consumption of computing resources, high calculation accuracy, and a relatively simple solution process. It provides an efficient solution for eliminating exogenous noise in vibration detection of wind tunnel vibration tests, and at the same time, the use of deep learning algorithms to solve modal analysis problems has a certain theoretical foresight. In short, the method for effectively eliminating multi-source mixed noise in wind tunnel experiments based on a deep learning algorithm proposed in this method is a relatively simple and practical method for general wind tunnel test scenarios. BRIEF DESCRIPTION OF THE DRAWINGS

[0047] Figure 1 It is a schematic diagram of the experimental system consisting of a force balance and an aircraft model;

[0048] Figure 2 It is a schematic diagram of dynamic simulation using the finite element method;

[0049] Figure 3 1 is a comparison diagram of the results of the method of the present invention, wherein (a) is a diagram of sampling data before noise removal, and (b) is a diagram of sampling data after noise removal;

[0050] Figure 4 is a comparison diagram of the results of the method of the present invention, wherein (a) is a schematic diagram of the reference data of the sampling points, and (b) is a schematic diagram of the relative error of the deep learning algorithm fitting;

[0051] Figure 5 This is a flow chart of an effective method for eliminating multi-source mixed noise in wind tunnel experiments based on a deep learning algorithm;

[0052] In the figure: 1-force measuring balance support rod, 2-aircraft model. DETAILED DESCRIPTION

[0053] The specific implementation of the present invention is further described below in conjunction with the accompanying drawings and technical solutions.

[0054] Example

[0055] A vibration system composed of a force balance and an aircraft model 2 is selected for modeling and calculation, wherein the force balance support rod 1 is mainly solved and analyzed.

[0056] An effective method for eliminating multi-source mixed noise in wind tunnel experiments based on deep learning algorithm Figure 5 As shown, the specific calculation steps are as follows:

[0057] Step 1: Establish a force balance buffeting estimation model

[0058] The force balance support rod 1 is assumed to be a homogeneous elastic rod with non-uniform cross-section, and its density is , take its Young's modulus as , take its length as , whose physical properties conform to various assumptions of classical elastic mechanics. Set the mass of aircraft model 2 to , the radius of the cross-section circle at the fixed position is First, we can derive the boundary conditions of the vibration differential equation

[0059]

[0060]

[0061] According to formula (2), the vibration estimation model of the force balance is established

[0062]

[0063] Step 2: Solve the vibration dynamics equation and perform modal analysis

[0064] According to formula (6), the solution of the vibration dynamics equation is:

[0065]

[0066] Select the modal order , to provide theoretical preparation for the deep learning algorithm in the third step.

[0067] Step 3: Vibration decoupling and noise removal based on deep learning algorithm

[0068] A GRU deep learning model is established, in which the number of GRU hidden layers is 2, the number of neurons is [3, 6], the hidden layer of the fully connected layer is 3 layers, the number of neurons is [12, 6, 3], and the number of neurons in the output layer is 1. Its estimated maximum error is 6.82×10-3g, the average absolute error is 9.20×10-4g, and the root mean square error is 1.2×10-3g. By fitting the measured point data, the numerical solutions of each mode of the vibration dynamics equation in the second step are determined. The results are as follows Figure 4 shown.

[0069] Substituting the solution of the deep learning model into equations (8) and (9), we can get the force noise and torque noise that need to be eliminated. Subtracting equation (6) from equation (8) and (9) can get the sampling data after eliminating the vibration noise. The result is as follows: Figure 3 shown.

[0070] At this point, the calculation of this embodiment is completed.

[0071] This method for effectively eliminating multi-source mixed noise in wind tunnel experiments based on deep learning algorithms is aimed at the common problems in vibration wind tunnel tests, such as limited test space, high frequency response requirements for measurement dimensions, mixed multi-source noise, and low precision of aerodynamic data. This method proposes a solution to effectively eliminate multi-source mixed noise based on deep learning algorithms. Starting from the vibration dynamics equation, this method performs vibration modal analysis, and gives the modal solution of the vibration dynamics equation based on the deep learning algorithm, thereby analyzing the mixed noise for elimination. It has the advantages of fast calculation speed, high precision, simple operation, and clear vibration modal analysis.

[0072] In addition to solving the embodiments proposed in this patent specification, this method can also use the idea of ​​using the vibration dynamics equations based on deep learning algorithms to propose modal solutions to solve other difficult problems in wind tunnel experiments. Compared with traditional methods, this method has certain flexibility and is a simple and efficient new method.

Claims

1. An effective method for eliminating multi-source mixed noise in wind tunnel experiments based on deep learning algorithm, characterized in that: Here are the steps: The first step is to establish a force balance buffeting estimation model; Consider an experimental system consisting of a force balance and an aircraft model. The two ends of the force balance support rod are fixedly connected to the force balance and the aircraft model respectively. The center of the circle corresponding to the tail cross section of the force balance support rod fixed to the force balance is taken as the origin of the coordinate system, and its axis is taken as Axis, the direction points from the origin to the opposite direction of the force balance support rod; According to the theory of elastic mechanics, the force balance support rod is equivalent to a variable cross-section elastic beam, and the differential equation of the forced vibration motion of the variable cross-section elastic beam is: Where: is the bending moment function, is the time term, is the density of the force balance rod, is the cross-sectional area function of the force balance, is the exciting force function, is the deflection function; According to the Euler-Bernoulli theory, the differential equation of the deflection vibration response of the variable cross-section beam is obtained, which is the buffeting estimation model of the force measuring balance: Where: is the Young's modulus of the force balance material, is the moment of inertia of the cross section of the force balance; At this point, the buffeting estimation model for the force balance has been established; The second step is to solve the vibration dynamics equation and perform modal analysis; According to the separation of variables method for solving partial differential equations, let Where: is the position-dependent function term in the deflection function, called the vibration mode function; is the time-dependent function term in the deflection function; say for The order natural vibration mode function satisfies the characteristic equation: Using the orthogonality of the vibration mode function, we get: Where: for The generalized force corresponding to the order vibration mode is, is the length of the variable cross-section elastic beam; Using Duhamel points, we get: Where: for The modal amplitude corresponding to the order natural vibration mode is, and They are The two constants under the order natural vibration mode are: for The angular frequency corresponding to the order natural vibration mode is, is the integral parameter, is a constant parameter; The natural vibration function called The modal shape function, each vibration mode corresponds to a vibration shape; At this point, the solution of the vibration dynamics equations and the modal analysis are completed; Step 3: Vibration decoupling and noise removal based on deep learning algorithm From formula (6), we know that the solution of the vibration mode generalized coordinate function is The order vibration mode and the corresponding integral constant are expressed. Therefore, based on the deep learning algorithm, according to the fitting function data of the acceleration measurement point in the experiment, the modal order to be determined is set to determine the integral constant. The relationship between the acceleration of the measurement point and the modal amplitude function is: Where: , Respectively Point and The acceleration at the point, represents the second-order derivative function of the fitted modal amplitude, and due to It is the fitting function of the experimental acceleration measurement point data, and is marked as an estimated quantity; A GRU deep learning model is established, where the number of hidden layers of GRU is 2, the number of neurons is [3, 6], the number of hidden layers of the fully connected layer is 3, the number of neurons is [12, 6, 3], and the number of neurons in the output layer is 1; its estimated maximum error is 6.82×10-3g, the average absolute error is 9.20×10-4g, and the root mean square error is 1.2×10-3g, where g represents the gravitational acceleration; the vibration acceleration at the force balance is estimated to achieve vibration decoupling, and the numerical solutions of each mode of equation (6) are determined by fitting the measurement point data; Due to the inertia and gravity of the aircraft model, a support reaction force will be generated on the force balance support rod, thereby generating a shear force in the lift direction on the force balance; since the center of mass of the aircraft model does not coincide with the measurement center, a support moment will be generated under the influence of gravity; and a pitch inertia moment will be generated during the oscillation of the aircraft model. Based on Newton's theorem and the parallel axis theorem, its function form is listed as follows: Where: , is the chattering noise value of force and torque, is the gravity acting on the aircraft model, is the angle between the aircraft model and the horizontal plane at the support point of the strut, is the inertial acceleration of the aircraft model, is the angular acceleration of the aircraft model, L1 is the distance from the center of gravity of the aircraft model to the origin of the coordinate system, L2 is the distance from the center of gravity of the aircraft model to the sensor, is the mass of the aircraft model; Combining equations (6), (7), (8), and (9) with the fitting results of the measurement point data obtained by the GRU deep learning model, the numerical solutions of the two vibration noises are obtained, which are then decoupled from the experimental data to obtain more accurate experimental results. At this point, vibration decoupling and noise removal based on deep learning algorithm are completed.

Images