Wind tunnel test data processing method suitable for rotating aircraft Magnus effect

Abstract

The invention belongs to the technical field of rotary aircraft wind tunnel tests, and particularly relates to a wind tunnel test data processing method suitable for the Magnus effect of a rotary aircraft, and the method comprises the following steps: 1, preprocessing an original voltage signal; step 2, calculation of aerodynamic load; step 3, carrying out time-frequency analysis processing on the transient signal; 4, averaging the transient value of each component to obtain a time-average aerodynamic coefficient; therefore, all data processing is completed, and the transient and time-average aerodynamic characteristics of the Magnus effect wind tunnel test of the rotating aircraft are finally obtained. According to the method, the transient balance signal is deeply processed and analyzed based on a conventional force measurement wind tunnel test data processing method, so that the transient aerodynamic law and time-average aerodynamic data of the rotating aircraft can be quickly and accurately obtained, and data support is provided for overall design iteration.

Description

Technical Field

[0001] The invention belongs to the technical field of wind tunnel tests on rotating aircraft, and in particular relates to a wind tunnel test data processing method suitable for the Magnus effect of a rotating aircraft. Background Art

[0002] The Magnus effect refers to the phenomenon in which a rotating aircraft, in flight at an angle of attack, generates lateral aerodynamic forces and yawing moments perpendicular to the plane of attack due to the interaction between its rotational motion and the incoming airflow. This phenomenon has a significant impact on the dynamic characteristics of the aircraft: although the magnitude of the generated Magnus force is relatively small, the accompanying yawing moment can cause the projectile axis to deviate from the plane of attack, severely impairing flight stability. This effect is the dominant factor in the dynamic instability of low-drag projectiles and also has a significant impact on the dynamic stability of low-speed rotating winged rockets.

[0003] Currently, the primary methods for studying the Magnus effect include numerical simulation and wind tunnel testing. However, because the generation of lateral force and yaw moment involves complex nonlinear flow mechanisms, relying solely on numerical simulation has significant limitations: large-scale grid generation is required for rotating aircraft of varying shapes, and each state point must be calculated individually using RANS methods or even large eddy simulations, resulting in enormous computational resource consumption and lengthy computation cycles. Therefore, in practical applications, wind tunnel testing remains the most effective method for studying the Magnus effect.

[0004] However, wind tunnel testing of the Magnus effect presents significant technical challenges. Due to the relatively small magnitude of the lateral force and yaw moment, and the susceptibility of transient balance signals to external interference, the quality of the obtained test data is insufficient to meet the desired requirements. This not only impacts test efficiency but also limits the accuracy of the research results. Therefore, there is an urgent need to develop an effective wind tunnel test data processing method to improve both test efficiency and data accuracy in research on the Magnus effect of rotating aircraft. Summary of the Invention

[0005] (1) Technical issues to be resolved

[0006] The technical problem to be solved by the present invention is: how to provide a wind tunnel test data processing method in order to improve the test efficiency and data accuracy in the wind tunnel test of the Magnus effect of a rotating aircraft.

[0007] (2) Technical solution

[0008] To solve the above technical problems, the present invention provides a method for processing wind tunnel test data applicable to the Magnus effect of a rotating aircraft, the method comprising the following steps:

[0009] Step 1: Preprocessing of the original voltage signal;

[0010] Step 2: Calculation of aerodynamic loads;

[0011] Step 3: Perform time-frequency analysis on the transient signal;

[0012] Step 4: average the transient values of each component to obtain the time-averaged aerodynamic coefficient;

[0013] At this point, all data processing has been completed, and finally the transient and time-averaged aerodynamic characteristics of the Magnus effect wind tunnel test of the rotating aircraft have been obtained.

[0014] Wherein, in the step 1, the original voltage signal is preprocessed, and the steps are as follows:

[0015] Step 1.1: Divide the balance signal collected during the test into steps according to the angle of attack to obtain the original transient balance signal matrix at different angles of attack;

[0016] Step 1.2: Process the digital signals collected by the photoelectric switch at different attack angles during the test to obtain the average speed value at each attack angle;

[0017] Step 1.3: Subtract the initial reading of the balance at each attack angle from the original transient balance signal matrix at different attack angles to obtain the incremental matrix of each balance signal.

[0018] In step 1.1, the original acquired balance signal sequence is split into the original transient balance signal matrix U at each attack angle according to the attack angle sequence. d ; Balance signal matrix U d There are fs·t rows and 5 columns; fs is the sampling frequency, t is the sampling time, and the 5 columns are the original balance signal voltage values U collected during the test. d1 、U d2 、U d3 、U d4 、U d5 , is the original transient balance signal matrix U d The 1st to 5th elements of ;

[0019] In step 1.2, the digital signals collected by the photoelectric switch under different attack angles during the test are processed to obtain the average speed value n under different attack angles. s , the calculation formula is (1):

[0020]

[0021] Where: Count end The last count value of the photoelectric switch, Count startis the first count value of the photoelectric switch, and t is the sampling time;

[0022] In step 1.3, the original balance signal matrix U in step (1.1) is d Subtract the initial reading U of the balance at each angle of attack c , where U c is the initial reading of the balance of the test system at each attack angle without wind; the incremental matrix U of the balance signal is obtained, which is expressed as (2):

[0023] U=U d -U c (2).

[0024] Wherein, in the said step 2, the aerodynamic load is calculated, and the steps are as follows:

[0025] Step 2.1, according to the balance working formula, calculate the balance signal increment matrix obtained in step 1.3 to obtain the aerodynamic load matrix under different attack angles;

[0026] Step 2.2: Based on the calculated aerodynamic load matrix at different attack angles, complete the elastic angle correction of the balance and support rod;

[0027] Step 2.3: Transfer the reference point of the aerodynamic load matrix corrected in step 2.2 from the calibration center to the warhead, and complete the dimensionless coefficient processing.

[0028] In step 2.1, the aerodynamic load matrix Y is calculated based on the incremental matrix U of the balance signal and the balance working formula. The matrix has fs·t rows and 5 columns, and the calculation expression is (3):

[0029]

[0030] Where: Y i is the measured value of the i-th component load of the balance; i = 1,…,5, representing the normal force, pitching moment, rolling moment, lateral force and yaw moment respectively, corresponding to the meaning of the five columns of voltage signal one by one; Y j 、Y k and Y i Similarly, they are the load measurement values of the jth and kth components of the balance, respectively; j = 1, ..., 5; k = 1, ..., 5;

[0031] a i is the main coefficient of the i-th component; b ij is the linear interference coefficient of the jth component load on the ith component; c ijk is the quadratic interference coefficient of the j, k component load on the i component; where a i 、b ij and c ijkAll given by the balance formula; U i is the voltage increment of the balance signal of the i-th component of the increment matrix U; i=1,…,5;

[0032] In step 2.2, the aerodynamic load matrix Y is used to correct the elastic angle of the balance and the support rod. The aerodynamic force and torque borne by the balance will cause the support rod to produce elastic deformation, resulting in a change in the posture of the model installed on the balance. In this test, since the lateral force and rolling moment are small, the changes in the sideslip angle and roll angle are ignored, and only the angle of attack is corrected. The calculation formula for the elastic deformation of the angle of attack is (4), and the calculation formula for the actual angle of attack is (5):

[0033]

[0034] α=α m +Δα e (5)

[0035] Where: is the angle of attack coefficient generated by the normal force; is the angle of attack coefficient generated by the pitching moment; α is the actual angle of attack, in degrees; Δα e is the elastic angle of the attack angle change, in degrees; α m is the nominal angle of attack, in degrees, given by the angle of attack mechanism in the test system; Y1 is the measured value of the first component load of the balance, and Y2 is the measured value of the second component load of the balance;

[0036] In step 2.3, since the reference point of the balance torque measurement is the balance calibration center, when the model torque reference point does not coincide with the balance calibration center, it is necessary to transfer the torque reference point measured by the balance to the model torque reference point; the corrected aerodynamic load matrix X is calculated as (6):

[0037]

[0038] Where: L x The axial distance between the moment reference point and the balance center, with the balance center in front being positive;

[0039] X1, X2, X3, X4, and X5 are the first to fifth elements in the aerodynamic load matrix X;

[0040] Y1, Y2, Y3, Y4, and Y5 are the load measurements of the 1st to 5th components of the balance;

[0041] Furthermore, the corrected aerodynamic load matrix X is dimensionless converted into the aerodynamic coefficient matrix C X , where the calculation expression of aerodynamic coefficient is (7), and the calculation expression of aerodynamic moment coefficient is (8),

[0042]

[0043] Where: q is the dynamic pressure of the flow field during the wind tunnel test, in Pa; S is the reference area of the test model, in m 2 ; L is the reference length of the test model, in m.

[0044] In step 3, time-frequency analysis is performed on the transient signal, and the steps are as follows:

[0045] Step 3.1: Perform weighted processing on the time domain signal using the Hamming window function to balance the resolution of the time domain and frequency domain;

[0046] Step 3.2, perform fast Fourier transform on the processed signal to obtain the complex spectrum and spectrogram;

[0047] Step 3.3: Based on the speed value calculated in step 1.2, design a bandpass filter whose main function is to extract frequencies 4 times and 8 times the speed;

[0048] Step 3.4: Complete the filtering operation based on the complex spectrum obtained in step 3.2 and the bandpass filter designed in step 3.3;

[0049] Step 3.5: Perform inverse Fourier transform on the filtered signal in step 3.4 to obtain the transient aerodynamic coefficient matrix at each angle of attack.

[0050] Among them, in step 3.1, in order to reduce spectrum leakage, the time domain signal aerodynamic coefficient matrix C is first X (t) Use Hamming window function to perform weighted processing to obtain the processed signal C X ′(t), where the length of the Hamming window is 2000 points to balance the resolution of the time domain and frequency domain;

[0051] Furthermore, in step 3.2, the windowed signal C X ′(t) is subjected to fast Fourier transform to obtain the complex spectrum C X ′(f), and draw the spectrum;

[0052] Furthermore, in step 3.3, a bandpass filter H(f) is designed; wherein the main frequency of the Magnus effect is the full-bomb rotation frequency n s 4 times and 8 times; therefore, the passband range of the bandpass filter is [4n s -2.5,4n s +2.5] and [8n s -2.5,8n s +2.5];

[0053] Furthermore, in step 3.4, signal filtering is performed in the frequency domain; the complex spectrum C in step 3.2 is converted to X ′(f) is multiplied by the bandpass filter H(f) in step 3.3 to obtain the filtered spectrum C Xh ′(f), complete the filtering operation; the expression is (9);

[0054] C Xh ′(f)=C X ′(f)·H(f) (9)

[0055] Furthermore, in step 3.5, the signal after filtering is subjected to inverse Fourier transform to restore it from the frequency domain to the time domain C Xh ′(t), thus completing the processing of transient test data and obtaining the transient aerodynamic coefficient matrix C under all states Xh ′.

[0056] In step 4, the time-averaged aerodynamic coefficients are obtained by averaging the transient values of the components. Thus, all data processing is completed, and the transient and time-averaged aerodynamic characteristics of the rotating aircraft Magnus effect wind tunnel test are finally obtained.

[0057] Among them, in the step 4, the transient signals of each angle of attack are averaged to obtain the time-averaged aerodynamic coefficients in all states, and finally the variation law of the Magnus effect of the rotating aircraft with the angle of attack and Mach number is obtained.

[0058] (3) Beneficial effects

[0059] Compared with the existing technology, the present invention is based on the conventional force measurement wind tunnel test data processing method, and deeply processes and analyzes the transient balance signal, so as to quickly and accurately obtain the transient aerodynamic laws and time-averaged aerodynamic data of the rotating aircraft, providing data support for the overall design iteration. BRIEF DESCRIPTION OF THE DRAWINGS

[0060] Figure 1 Diagram of the model and balance setup for the Magnus effect wind tunnel test.

[0061] Figure 2 This is the digital signal diagram of the photoelectric switch (alfa = 12°).

[0062] Figure 3 The graph of the untreated lateral force coefficient over time (alfa = 12°).

[0063] Figure 4 Spectrum diagram of the lateral force coefficient (alfa = 12°).

[0064] Figure 5The graph of the lateral force coefficient after treatment with time (alfa = 12°).

[0065] Figure 6 This is a graph showing the variation of the time-averaged total missile side force coefficient with the attack angle. DETAILED DESCRIPTION

[0066] In order to make the purpose, content, and advantages of the present invention more clear, the specific implementation methods of the present invention are further described in detail below with reference to the accompanying drawings and examples.

[0067] To solve the above technical problems, the present invention provides a method for processing wind tunnel test data applicable to the Magnus effect of a rotating aircraft, the method comprising the following steps:

[0068] Step 1: Preprocessing of the original voltage signal;

[0069] Step 2: Calculation of aerodynamic loads;

[0070] Step 3: Perform time-frequency analysis on the transient signal;

[0071] Step 4: average the transient values of each component to obtain the time-averaged aerodynamic coefficient;

[0072] At this point, all data processing has been completed, and finally the transient and time-averaged aerodynamic characteristics of the Magnus effect wind tunnel test of the rotating aircraft have been obtained.

[0073] Wherein, in the step 1, the original voltage signal is preprocessed, and the steps are as follows:

[0074] Step 1.1: Divide the balance signal collected during the test into steps according to the angle of attack to obtain the original transient balance signal matrix at different angles of attack;

[0075] Step 1.2: Process the digital signals collected by the photoelectric switch at different attack angles during the test to obtain the average speed value at each attack angle;

[0076] Step 1.3: Subtract the initial reading of the balance at each attack angle from the original transient balance signal matrix at different attack angles to obtain the incremental matrix of each balance signal.

[0077] In step 1.1, the original acquired balance signal sequence is split into the original transient balance signal matrix U at each attack angle according to the attack angle sequence. d ; Balance signal matrix U d There are fs·t rows and 5 columns; fs is the sampling frequency, t is the sampling time, and the 5 columns are the original balance signal voltage values U collected during the test. d1 、U d2 、U d3 、Ud4 、U d5 , is the original transient balance signal matrix U d The 1st to 5th elements of ;

[0078] In step 1.2, the digital signals collected by the photoelectric switch under different attack angles during the test are processed to obtain the average speed value n under different attack angles. s , the calculation formula is (1):

[0079]

[0080] Where: Count end The last count value of the photoelectric switch, Count start is the first count value of the photoelectric switch, and t is the sampling time;

[0081] In step 1.3, the original balance signal matrix U in step (1.1) is d Subtract the initial reading U of the balance at each angle of attack c , where U c is the initial reading of the balance of the test system at each attack angle without wind; the incremental matrix U of the balance signal is obtained, which is expressed as (2):

[0082] U=U d -U c (2).

[0083] Wherein, in the said step 2, the aerodynamic load is calculated, and the steps are as follows:

[0084] Step 2.1, according to the balance working formula, calculate the balance signal increment matrix obtained in step 1.3 to obtain the aerodynamic load matrix under different attack angles;

[0085] Step 2.2: Based on the calculated aerodynamic load matrix at different attack angles, complete the elastic angle correction of the balance and support rod;

[0086] Step 2.3: Transfer the reference point of the aerodynamic load matrix corrected in step 2.2 from the calibration center to the warhead, and complete the dimensionless coefficient processing.

[0087] In step 2.1, the aerodynamic load matrix Y is calculated based on the incremental matrix U of the balance signal and the balance working formula. The matrix has fs·t rows and 5 columns, and the calculation expression is (3):

[0088]

[0089] Where: Y iis the measured value of the i-th component load of the balance; i = 1,…,5, representing the normal force, pitching moment, rolling moment, lateral force and yaw moment respectively, corresponding to the meaning of the five columns of voltage signal one by one; Y j 、Y k and Y i Similarly, they are the load measurement values of the jth and kth components of the balance, respectively; j = 1, ..., 5; k = 1, ..., 5;

[0090] a i is the main coefficient of the i-th component; b ij is the linear interference coefficient of the jth component load on the ith component; c ijk is the quadratic interference coefficient of the j, k component load on the i component; where a i 、b ij and c ijk All given by the balance formula; U i is the voltage increment of the balance signal of the i-th component of the increment matrix U; i=1,…,5;

[0091] In step 2.2, the aerodynamic load matrix Y is used to correct the elastic angle of the balance and the support rod. The aerodynamic force and torque borne by the balance will cause the support rod to produce elastic deformation, resulting in a change in the posture of the model installed on the balance. In this test, since the lateral force and rolling moment are small, the changes in the sideslip angle and roll angle are ignored, and only the angle of attack is corrected. The calculation formula for the elastic deformation of the angle of attack is (4), and the calculation formula for the actual angle of attack is (5):

[0092]

[0093]

[0094] Where: is the angle of attack coefficient generated by the normal force; is the angle of attack coefficient generated by the pitching moment; α is the actual angle of attack, in degrees; Δα e is the elastic angle of the attack angle change, in degrees; α m is the nominal angle of attack, in degrees, given by the angle of attack mechanism in the test system; Y1 is the measured value of the first component load of the balance, and Y2 is the measured value of the second component load of the balance;

[0095] In step 2.3, since the reference point of the balance torque measurement is the balance calibration center, when the model torque reference point does not coincide with the balance calibration center, it is necessary to transfer the torque reference point measured by the balance to the model torque reference point; the corrected aerodynamic load matrix X is calculated as (6):

[0096]

[0097] Where: Lx The axial distance between the moment reference point and the balance center, with the balance center in front being positive;

[0098] X1, X2, X3, X4, and X5 are the first to fifth elements in the aerodynamic load matrix X;

[0099] Y1, Y2, Y3, Y4, and Y5 are the load measurements of the 1st to 5th components of the balance;

[0100] Furthermore, the corrected aerodynamic load matrix X is dimensionless converted into the aerodynamic coefficient matrix C X , where the calculation expression of aerodynamic coefficient is (7), and the calculation expression of aerodynamic moment coefficient is (8),

[0101]

[0102] Where: q is the dynamic pressure of the flow field during the wind tunnel test, in Pa; S is the reference area of the test model, in m 2 ; L is the reference length of the test model, in m.

[0103] In step 3, time-frequency analysis is performed on the transient signal, and the steps are as follows:

[0104] Step 3.1: Perform weighted processing on the time domain signal using the Hamming window function to balance the resolution of the time domain and frequency domain;

[0105] Step 3.2, perform fast Fourier transform on the processed signal to obtain the complex spectrum and spectrogram;

[0106] Step 3.3: Based on the speed value calculated in step 1.2, design a bandpass filter whose main function is to extract frequencies 4 times and 8 times the speed;

[0107] Step 3.4: Complete the filtering operation based on the complex spectrum obtained in step 3.2 and the bandpass filter designed in step 3.3;

[0108] Step 3.5: Perform inverse Fourier transform on the filtered signal in step 3.4 to obtain the transient aerodynamic coefficient matrix at each angle of attack.

[0109] Among them, in step 3.1, in order to reduce spectrum leakage, the time domain signal aerodynamic coefficient matrix C is first X (t) Use Hamming window function to perform weighted processing to obtain the processed signal C X ′(t), where the length of the Hamming window is 2000 points to balance the resolution of the time domain and frequency domain;

[0110] Furthermore, in step 3.2, the windowed signal C X ′(t) is subjected to fast Fourier transform to obtain the complex spectrum C X ′(f), and draw the spectrum;

[0111] Furthermore, in step 3.3, a bandpass filter H(f) is designed; wherein the main frequency of the Magnus effect is the full-bomb rotation frequency n s 4 times and 8 times; therefore, the passband range of the bandpass filter is [4n s -2.5,4n s +2.5] and [8n s -2.5,8n s +2.5];

[0112] Furthermore, in step 3.4, signal filtering is performed in the frequency domain; the complex spectrum C in step 3.2 is converted to X ′(f) is multiplied by the bandpass filter H(f) in step 3.3 to obtain the filtered spectrum C Xh ′(f), complete the filtering operation; the expression is (9);

[0113] C Xh ′(f)=C X ′(f)·H(f) (9)

[0114] Furthermore, in step 3.5, the signal after filtering is subjected to inverse Fourier transform to restore it from the frequency domain to the time domain C Xh ′(t), thus completing the processing of transient test data and obtaining the transient aerodynamic coefficient matrix C under all states Xh ′.

[0115] In step 4, the time-averaged aerodynamic coefficients are obtained by averaging the transient values of the components. Thus, all data processing is completed, and the transient and time-averaged aerodynamic characteristics of the rotating aircraft Magnus effect wind tunnel test are finally obtained.

[0116] Among them, in the step 4, the transient signals of each angle of attack are averaged to obtain the time-averaged aerodynamic coefficients in all states, and finally the variation law of the Magnus effect of the rotating aircraft with the angle of attack and Mach number is obtained.

[0117] Example 1

[0118] This embodiment, in order to improve test efficiency and data accuracy during a Magnus effect wind tunnel test of a rotating aircraft, deeply processes and analyzes transient balance signals based on conventional force measurement wind tunnel test data processing methods. This allows for rapid and accurate acquisition of transient aerodynamic patterns and time-averaged aerodynamic data, providing data support for overall design iterations. Specifically, the present invention discloses a data processing method for a Magnus effect wind tunnel test of a rotating aircraft, comprising the following steps:

[0119] Step 1: Preprocess the original balance voltage signal.

[0120] The Magnus effect test of the rotating aircraft was carried out in the CG-01 high-speed wind tunnel. The test model was the basic-finner standard model. The model and the force balance were as follows. Figure 1 As shown in the figure. During the test, the Mach number Ma = 2.5, and the angles of attack are 0°, 2°, 4°, 6°, 8°, 10°, 12°, and 14°. The data sampling frequency fs = 5000 Hz, and the sampling time t = 5s at each angle of attack. The original collected balance voltage signal is split into the original transient balance signal matrix U at each angle of attack according to the angle of attack sequence. d , where the matrix has a total of fs·t=25000 rows and 5 columns, the 5 columns are respectively the collected U y 、U mz 、U mx 、U z and U my Due to the large amount of data, the subsequent transient signal processing is only based on the attack angle of 12°. The data processing methods for other attack angles are the same.

[0121] The digital signals collected by the photoelectric switch at different attack angles during the test are processed to obtain the speed values at different attack angles. When the attack angle is 12°, the digital signals collected by the photoelectric switch are as follows: Figure 2 As shown. Then according to expression (1), the speed value n at this angle of attack is calculated s The same method is used to process the other angles of attack.

[0122]

[0123] Then, the initial reading of the collected transient balance signal is deducted to obtain the voltage signal increment under the actual aerodynamic load. d Subtract the initial reading U of the balance without wind at each angle of attack c , thereby obtaining the incremental matrix U of the balance signal, the expression is (2):

[0124] U=U d -U c (2)

[0125] Step 2: Calculate the aerodynamic loads under each state.

[0126] In the previous step, the balance signal increments at different angles of attack were obtained. Based on the signal increments, the aerodynamic load matrix Y at each angle of attack was calculated. This matrix has the same structure as the balance signal increment matrix U, with a total of 25,000 rows and 5 columns. The calculation expression is (3):

[0127]

[0128] Where: Y i is the measured value of the i-th component load of the balance; i = 1,…,5, representing the normal force, pitching moment, rolling moment, lateral force and yaw moment respectively, and corresponding to the five columns of the voltage signal one by one; a i is the main coefficient of the i-th component; b ij is the linear interference coefficient of the jth component load on the ith component; c ijk is the quadratic interference coefficient of the j, k component load on the i component; where a i 、b ij and c ijk All are given by the balance formula.

[0129] During the test, when the test model is subjected to aerodynamic loads, the force balance and support rods will produce a certain amount of elastic deformation, causing the actual posture of the model to change. In this test, since the lateral force and rolling moment are small, the changes in the sideslip angle and roll angle are ignored, and only the angle of attack is corrected. The formula for calculating the elastic angle is (4), and the formula for calculating the actual angle of attack is (5):

[0130]

[0131] Where: is the angle of attack coefficient generated by the normal force, is the angle of attack coefficient generated by the pitching moment.

[0132] α=α m +Δα e (5)

[0133] Where: α is the actual angle of attack; α m is the nominal angle of attack, which is given by the angle of attack mechanism in the test system.

[0134] In the test, the torque reference point measured by the balance is the calibration center of the balance. Generally, the torque reference point given by the wind tunnel test is the model head. Therefore, it is necessary to change the torque reference point measured by the balance to the model head. The corrected aerodynamic load matrix X is calculated as (6):

[0135]

[0136] Where: L x The axial distance between the moment reference point and the balance center, with the balance center in front being positive;

[0137] Finally, the corrected aerodynamic load matrix X is dimensionless converted into the aerodynamic coefficient matrix C X , where the aerodynamic force calculation expression is (7), and the aerodynamic moment calculation expression is (8),

[0138]

[0139] Where: q is the dynamic pressure of the flow field during the wind tunnel test, in Pa; S is the reference area of the test model, in m 2 ; L is the reference length of the test model, in m.

[0140] Step 3: Perform time-frequency analysis on the transient signal.

[0141] In the previous step, the aerodynamic coefficient matrix C of the model at each angle of attack was obtained. X , the matrix is the signal in the time domain. When the angle of attack is 12°, the signal of the lateral force coefficient in the time domain is as follows Figure 3 As shown (only data within 0.2s are given). In order to remove the noise signal in the experiment from the transient signal, the required signal is extracted and time-frequency analysis is performed on the transient signal in this step.

[0142] In order to reduce spectrum leakage, the aerodynamic coefficient matrix C of the time domain signal is first X Each component signal in (t) is weighted using the Hamming window function to obtain the processed signal C X ′(t), where the length of the Hamming window is 2000 points to balance the resolution of the time domain and the frequency domain.

[0143] After that, the windowed signal C X ′(t) is subjected to fast Fourier transform to obtain the complex spectrum C X ′(f), and draw the spectrum. Taking the side force coefficient at an attack angle of 12° as an example, its signal in the time domain and its spectrum in the frequency domain are as follows: Figure 4 As shown in the figure, the main frequencies of the signal are 463Hz and 926Hz, which are 4 times and 8 times the full-bomb rotational speed frequency.

[0144] Then, according to the characteristics of the spectrum, a bandpass filter H(f) is designed. Its main function is to extract the 4th and 8th times of the full-bomb rotational speed frequency. Therefore, the passband range of the bandpass filter is [4n s -2.5,4n s +2.5] and [8ns -2.5,8n s +2.5].

[0145] After that, the signal is filtered in the frequency domain. The complex spectrum C X ′(f) is multiplied by the bandpass filter H(f) to obtain the filtered spectrum C Xh ′(f), the filtering operation is completed. The expression is (9).

[0146] C Xh ′(f)=C X ′(f)·H(f) (9)

[0147] Finally, for the filtered signal C Xh ′(f) performs inverse Fourier transform to restore it from the frequency domain to the time domain C Xh ′(t), thus completing the processing of transient test data and obtaining the transient aerodynamic coefficient matrix C under all states Xh ′. When the angle of attack is 12°, the side force coefficient after processing changes with time as shown in the following figure: Figure 5 shown.

[0148] Step 4: Average the transient values of each component to obtain the time-averaged value.

[0149] After completing the time-frequency analysis of the aerodynamic data at all attack angles, the transient signals in each state are averaged to obtain the time-averaged aerodynamic coefficients in all states, and finally the variation law of the Magnus effect of the rotating aircraft with the attack angle is obtained. The time-averaged total missile side force coefficient changes as shown in the figure below: Figure 6 shown.

[0150] The above is only a preferred embodiment of the present invention. It should be pointed out that for ordinary technicians in this technical field, several improvements and modifications can be made without departing from the technical principles of the present invention. These improvements and modifications should also be regarded as the scope of protection of the present invention.

Claims

1. A method for processing wind tunnel test data for the Magnus effect of a rotating aircraft, characterized in that: The method comprises the following steps: Step 1: Preprocessing of the original voltage signal; Step 2: Calculation of aerodynamic loads; Step 3: Perform time-frequency analysis on the transient signal; Step 4: average the transient values of each component to obtain the time-averaged aerodynamic coefficient; At this point, all data processing has been completed, and finally the transient and time-averaged aerodynamic characteristics of the Magnus effect wind tunnel test of the rotating aircraft have been obtained.

2. The method for processing wind tunnel test data for the Magnus effect of a rotating aircraft according to claim 1, characterized in that: In step 1, the original voltage signal is preprocessed as follows: Step 1.1: Divide the balance signal collected during the test into steps according to the angle of attack to obtain the original transient balance signal matrix at different angles of attack; Step 1.2: Process the digital signals collected by the photoelectric switch at different attack angles during the test to obtain the average speed value at each attack angle; Step 1.3: Subtract the initial reading of the balance at each attack angle from the original transient balance signal matrix at different attack angles to obtain the incremental matrix of each balance signal.

3. The method for processing wind tunnel test data applicable to the Magnus effect of a rotating aircraft according to claim 2, characterized in that: In step 1.1, the original acquired balance signal sequence is split into the original transient balance signal matrix U at each attack angle according to the attack angle sequence. d ; Balance signal matrix U d There are fs·t rows and 5 columns; fs is the sampling frequency, t is the sampling time, and the 5 columns are the original balance signal voltage values U collected during the test. d1 、U d2 、U d3 、U d4 、U d5 , is the original transient balance signal matrix U d The 1st to 5th elements of ; In step 1.2, the digital signals collected by the photoelectric switch under different attack angles during the test are processed to obtain the average speed value n under different attack angles. s , the calculation formula is (1): Where: Count end The last count value of the photoelectric switch, Count start is the first count value of the photoelectric switch, and t is the sampling time; In step 1.3, the original balance signal matrix U in step (1.1) is d Subtract the initial reading U of the balance at each angle of attack c , where U c is the initial reading of the balance of the test system at each attack angle without wind; the incremental matrix U of the balance signal is obtained, which is expressed as (2): U=U d -IN c (2)。 4. The method for processing wind tunnel test data for the Magnus effect of a rotating aircraft according to claim 3, wherein: In step 2, the aerodynamic load is calculated as follows: Step 2.1, according to the balance working formula, calculate the balance signal increment matrix obtained in step 1.3 to obtain the aerodynamic load matrix under different attack angles; Step 2.2: Based on the calculated aerodynamic load matrix at different attack angles, complete the elastic angle correction of the balance and support rod; Step 2.3: Transfer the reference point of the aerodynamic load matrix corrected in step 2.2 from the calibration center to the warhead, and complete the dimensionless coefficient processing.

5. The method for processing wind tunnel test data applicable to the Magnus effect of a rotating aircraft according to claim 4, characterized in that: In step 2.1, the aerodynamic load matrix Y is calculated based on the incremental matrix U of the balance signal and the balance working formula. The matrix has fs·t rows and 5 columns, and the calculation expression is (3): Where: Y i is the measured value of the i-th component load of the balance; i = 1,…,5, representing the normal force, pitching moment, rolling moment, lateral force and yaw moment respectively, corresponding to the meaning of the five columns of voltage signal one by one; Y j 、Y k and Y i Similarly, they are the load measurement values of the jth and kth components of the balance, respectively; j = 1, ..., 5; k = 1, ..., 5; a i is the main coefficient of the i-th component; b ij is the linear interference coefficient of the jth component load on the ith component; c ijk is the quadratic interference coefficient of the j, k component load on the i component; where a i 、b ij and c ijk All given by the balance formula; U i is the voltage increment of the balance signal of the i-th component of the increment matrix U; i=1,…,5; In step 2.2, the aerodynamic load matrix Y is used to correct the elastic angle of the balance and the support rod. The aerodynamic force and torque borne by the balance will cause the support rod to produce elastic deformation, resulting in a change in the posture of the model installed on the balance. In this test, since the lateral force and rolling moment are small, the changes in the sideslip angle and roll angle are ignored, and only the angle of attack is corrected. The calculation formula for the elastic deformation of the angle of attack is (4), and the calculation formula for the actual angle of attack is (5): a = a m +Da e (5) Where: is the angle of attack coefficient generated by the normal force; is the angle of attack coefficient generated by the pitching moment; α is the actual angle of attack, in degrees; Δα e is the elastic angle of the attack angle change, in degrees; α m is the nominal angle of attack, in degrees, given by the angle of attack mechanism in the test system; Y1 is the measured value of the first component load of the balance, and Y2 is the measured value of the second component load of the balance; In step 2.3, since the reference point of the balance torque measurement is the balance calibration center, when the model torque reference point does not coincide with the balance calibration center, it is necessary to transfer the torque reference point measured by the balance to the model torque reference point; the corrected aerodynamic load matrix X is calculated as (6): Where: L x The axial distance between the moment reference point and the balance center, with the balance center in front being positive; X1, X2, X3, X4, and X5 are the first to fifth elements in the aerodynamic load matrix X; Y1, Y2, Y3, Y4, and Y5 are the load measurements of the 1st to 5th components of the balance; Furthermore, the corrected aerodynamic load matrix X is dimensionless converted into the aerodynamic coefficient matrix C X , where the calculation expression of aerodynamic coefficient is (7), and the calculation expression of aerodynamic moment coefficient is (8), Where: q is the dynamic pressure of the flow field during the wind tunnel test, in Pa; S is the reference area of the test model, in m 2 ; L is the reference length of the test model, in m.

6. The method for processing wind tunnel test data applicable to the Magnus effect of a rotating aircraft according to claim 5, characterized in that: In step 3, time-frequency analysis is performed on the transient signal, and the steps are as follows: Step 3.1: Perform weighted processing on the time domain signal using the Hamming window function to balance the resolution of the time domain and frequency domain; Step 3.2, perform fast Fourier transform on the processed signal to obtain the complex spectrum and spectrogram; Step 3.3: Based on the speed value calculated in step 1.2, design a bandpass filter whose main function is to extract frequencies 4 times and 8 times the speed; Step 3.4: Complete the filtering operation based on the complex spectrum obtained in step 3.2 and the bandpass filter designed in step 3.3; Step 3.5: Perform inverse Fourier transform on the filtered signal in step 3.4 to obtain the transient aerodynamic coefficient matrix at each angle of attack.

7. The method for processing wind tunnel test data applicable to the Magnus effect of a rotating aircraft according to claim 6, characterized in that: In step 3.1, in order to reduce spectrum leakage, the time domain signal aerodynamic coefficient matrix C is first X (t) Use Hamming window function to perform weighted processing to obtain the processed signal C X ′(t), where the length of the Hamming window is 2000 points to balance the resolution of the time domain and frequency domain; Furthermore, in step 3.2, the windowed signal C X ′(t) is subjected to fast Fourier transform to obtain the complex spectrum C X ′(f), and draw the spectrum; Furthermore, in step 3.3, a bandpass filter H(f) is designed; wherein the main frequency of the Magnus effect is the full-bomb rotation frequency n s 4 times and 8 times; therefore, the passband range of the bandpass filter is [4n s -2.5,4n s +2.5] and [8n s -2.5,8n s +2.5]; Furthermore, in step 3.4, signal filtering is performed in the frequency domain; the complex spectrum C in step 3.2 is converted to X ′(f) is multiplied by the bandpass filter H(f) in step 3.3 to obtain the filtered spectrum C Xh ′(f), complete the filtering operation; the expression is (9); C Xh ′(f)=C X ′(f)·H(f) (9) Furthermore, in step 3.5, the signal after filtering is subjected to inverse Fourier transform to restore it from the frequency domain to the time domain C Xh ′(t), thus completing the processing of transient test data and obtaining the transient aerodynamic coefficient matrix C under all states Xh ′.

8. The method for processing wind tunnel test data applicable to the Magnus effect of a rotating aircraft according to claim 7, characterized in that: In step 4, the time-averaged aerodynamic coefficient is obtained by averaging the transient values of each component; thus, all data processing is completed, and finally the transient and time-averaged aerodynamic characteristics of the rotating aircraft Magnus effect wind tunnel test are obtained.

9. The method for processing wind tunnel test data applicable to the Magnus effect of a rotating aircraft according to claim 8, characterized in that: In step 4, the transient signals at each angle of attack are averaged to obtain the time-averaged aerodynamic coefficients under all states, and finally the variation pattern of the Magnus effect of the rotating aircraft with the angle of attack and the Mach number is obtained.

10. The method for processing wind tunnel test data applicable to the Magnus effect of a rotating aircraft according to claim 9, characterized in that: The method is based on the conventional force measurement wind tunnel test data processing method, and deeply processes and analyzes the transient balance signal, so as to quickly and accurately obtain the transient aerodynamic laws and time-averaged aerodynamic data of the rotating aircraft, providing data support for the overall design iteration.

Images