A multi-source data fusion low-speed wind tunnel buffet test method and device

Abstract

The application discloses a kind of low-speed wind tunnel buffet test method and equipment of multi-source data fusion, belong to low-speed wind tunnel buffet test technical field.Solve the problem of low precision of test in the prior art traditional wind tunnel buffet test method and equipment multi-source data fusion ability insufficient;The application obtains test model sensing data and environmental parameters by the multi-source data synchronous acquisition of data acquisition module;Zero point calibration and signal conversion are carried out by data preprocessing module, and converted data is obtained;Multi-dimensional parameter collaborative calculation is carried out by multi-source fusion calculation module, and parameters are calculated;Determine the buffet starting angle of attack;Buffet coefficient is calculated, and buffet grade is determined, buffet characteristic curve is generated, and data output is completed.The application effectively improves the precision of buffet test, and can be applied to the buffet boundary identification, aerodynamic load characteristic test, structure response analysis and buffet grade evaluation of various general aircraft in low-speed high angle of attack state.

Description

Technical Field

[0001] This invention relates to the field of low-speed wind tunnel buffeting testing technology, specifically to a low-speed wind tunnel buffeting testing method and equipment based on multi-source data fusion. Background Technology

[0002] Flutter is a forced vibration of the aircraft structure caused by airflow separation at locations such as the nose and wings. Its excitation often originates from the separated flow on the main wing. As one of the key aeroelastic issues to consider during aircraft design, flutter directly affects the aircraft's flight safety, comfort, and structural durability. Low-speed wind tunnels, as the core testing platform for simulating low-speed flight environments (such as takeoff and landing), provide crucial data for aircraft aerodynamic design, structural optimization, and the development of flutter suppression measures. They are widely used in the research and development of various types of aircraft, including civil aircraft, drones, and low-altitude economic aircraft.

[0003] Currently, the core requirement of low-speed wind tunnel buffeting tests is to accurately acquire the buffeting characteristic parameters of aircraft models under low-speed flow field environments. These parameters include lift and axial force, pulsating pressure, structural vibration response, and flow field velocity distribution, thereby revealing the generation mechanism, evolution law, and influencing factors of buffeting. Existing low-speed wind tunnel buffeting test methods mostly collect test data through a single type of testing equipment. For example, they use a balance to obtain lift and axial force, a pulsating pressure sensor to obtain the pressure pulsation signal on the model surface, an accelerometer to collect the structural vibration response, and a flow field velocimeter to obtain flow field parameters, and then perform buffeting characteristic analysis based on a single data source.

[0004] As aircraft design evolves towards higher precision and reliability, traditional low-speed wind tunnel buffeting testing methods based on a single data source have gradually revealed numerous limitations, failing to meet the actual needs of engineering research and development. First, single data sources have limited information dimensions. Buffeting is the result of the coupled effects of multiple factors, including flow field, structural vibration, and aerodynamic loads. Relying solely on one type of data cannot comprehensively reflect the complete characteristics of buffeting. For example, collecting only pressure data makes it difficult to correlate structural vibration with changes in the flow field; monitoring only the vibration response cannot accurately identify the excitation source of buffeting, easily leading to deviations in buffeting mechanism analysis and affecting the reliability of experimental conclusions. Second, existing experimental data is susceptible to interference and lacks precision. During low-speed wind tunnel testing, factors such as tunnel wall interference, support interference, and environmental noise can cause errors in single data sources. Traditional methods lack effective data correction mechanisms, making it difficult to eliminate interference signals, thus affecting the measurement accuracy of buffeting parameters and failing to provide precise data support for aircraft design.

[0005] To address the limitations of single data sources, researchers have attempted to introduce multi-source data fusion technology to integrate and analyze experimental data from different types and sources, thereby improving the accuracy and comprehensiveness of buffeting tests. By complementing and fusing data from different testing equipment and different measurement dimensions, multi-source data fusion technology can effectively compensate for the deficiencies of a single data source, uncover the inherent correlations between data, and improve data reliability and information utilization. It has already been initially applied in fields such as aerodynamic modeling and wind tunnel test data processing. However, the application of existing multi-source data fusion technology in low-speed wind tunnel buffeting tests still has significant shortcomings, and a mature and systematic test method and supporting equipment have not yet been formed. On the one hand, existing fusion methods are mostly for specific types of data (such as only fusion of pressure and vibration data), lacking the ability to comprehensively fuse multi-dimensional and multi-type data (lift, axial force, pressure, vibration, flow field, aerodynamic load, etc.). Moreover, the nonlinear characteristics of the low-speed wind tunnel flow field and the heterogeneity of the data are not fully considered during the fusion process, which easily leads to problems such as data redundancy and fusion deviation, making it impossible to achieve accurate characterization of buffeting characteristics. On the other hand, existing test equipment lacks targeted multi-source data synchronous acquisition, transmission, and fusion processing modules. The data acquisition frequency and format of different test equipment are not uniform, which makes it difficult to achieve synchronous fusion of multi-source data and results in low data processing efficiency, failing to meet the needs of real-time analysis and real-time adjustment during the test, thus limiting the application effect of multi-source data fusion technology in low-speed wind tunnel buffeting tests. Furthermore, existing low-speed wind tunnel buffeting testing methods suffer from cumbersome procedures, low automation, and long testing cycles. The testing process requires manual data collection, processing, and analysis, increasing the workload of testing personnel and introducing human error, further impacting accuracy and efficiency. Meanwhile, research on buffeting tests for civil aircraft at low speeds and high angles of attack is relatively limited. Existing methods are insufficient to meet the buffeting characteristic testing requirements during critical phases such as takeoff and landing, failing to provide effective technical support for buffeting suppression design of these aircraft.

[0006] In summary, current low-speed wind tunnel buffeting testing methods and equipment suffer from technical deficiencies such as limited data acquisition dimensions, insufficient multi-source data fusion capabilities, low testing accuracy, low automation, and poor adaptability. These limitations make it difficult to meet the high-precision, comprehensive, and efficient requirements of modern aircraft development for buffeting testing. Therefore, a low-speed wind tunnel buffeting testing method and equipment based on multi-source data fusion is needed. Summary of the Invention

[0007] A brief overview of the invention is given below to provide a basic understanding of certain aspects of it. It should be understood that this overview is not an exhaustive summary of the invention. It is not intended to identify key or essential parts of the invention, nor is it intended to limit the scope of the invention. Its purpose is merely to present certain concepts in a simplified form as a prelude to the more detailed description that follows.

[0008] In view of this, in order to solve the problems of insufficient multi-source data fusion capability and low test accuracy of traditional wind tunnel buffeting test methods and equipment in the prior art, the present invention provides a low-speed wind tunnel buffeting test method and equipment with multi-source data fusion.

[0009] Technical solution one is as follows: A low-speed wind tunnel buffeting test method based on multi-source data fusion, comprising the following steps:

[0010] S1. Multi-source data is collected synchronously through the data acquisition module to obtain sensor data and environmental parameters of the test model;

[0011] S2. Zero-point calibration is performed on the sensor data of the test model under windless load conditions through the data preprocessing module. Signal conversion is performed on the obtained zero-point data and the sensor data of the test model to obtain the converted data.

[0012] S3. Based on the parameters obtained in steps S1 and S2, multi-dimensional parameter collaborative calculation is performed through the multi-source fusion calculation module to calculate the pulsating pressure parameters, structural response parameters, power spectral density and derived parameters.

[0013] S4. Combining the parameters obtained in steps S1-S3, determine the initial angle of attack for fluttering using the pulsating pressure method, wingtip acceleration method, wing root bending moment method, or steady aerodynamic curve inflection point method.

[0014] S5. Based on the determined initial angle of attack of the buffeting, calculate the buffeting coefficient, determine the buffeting level, automatically associate the attitude data of continuous angle of attack / sideslip angle, generate the buffeting characteristic curve, and complete the data output according to the standard format of aviation test.

[0015] Furthermore, in step S1, a pulsating pressure sensor array is arranged on the surface of the test model wing, strain gauges are attached to key stress locations at the wing root, accelerometers are arranged at the wingtip, and a force balance is installed at the wing support location. This completes the arrangement of the data acquisition module, which synchronously acquires sensor data from the test model, i.e., the pulsating pressure signal on the surface of the test model. Voltage signal of the strain gauge at the wing root wingtip accelerometer signal And simultaneously collect wind tunnel velocity and pressure Incoming static pressure Angle of attack Sideslip angle Speed ​​of sound Atmospheric pressure Reference sound pressure Including environmental parameters such as temperature T, ensure that all signals are triggered by the same clock and sampled using the same time base.

[0016] Furthermore, in step S2, baseline pressure data under windless conditions is collected. According to the incoming static pressure Zero-point data were calculated. ;

[0017] Zero point data Represented as

[0018]

[0019] in, To perform an averaging operation;

[0020] Zero-point data is converted through pulsed pressure conversion. and the pulsating pressure signal on the surface of the test model The data is converted into a pressure value, which is the converted pressure value obtained from the pulsating pressure sensor during the experiment.

[0021] Converted pressure value Represented as:

[0022]

[0023] Where a is the coefficient of the pulsating pressure sensor;

[0024] The DC offset and noise are removed from the time-domain signal to obtain the converted acceleration signal.

[0025] Furthermore, step S3 includes the following steps:

[0026] S31. Calculate the total sound pressure level based on the sensor data of the test model and the converted data;

[0027] In step S31, the local static pressure is obtained by averaging the converted pressure values. Take the converted pressure value and local static pressure The difference in value is obtained from the pulsating pressure. Further solving yields the root mean square of the pulsating pressure. Combined with wind tunnel speed pressure Solving for the root mean square of the pulsating pressure coefficient yields the solution. Combined with reference sound pressure Solving for the total sound pressure level (OASPL), we obtain the total sound pressure level (OASPL).

[0028] Local static pressure Represented as:

[0029] =mean( )

[0030] Pulsating pressure Represented as:

[0031]

[0032] Root mean square of pulsating pressure Represented as:

[0033]

[0034] Root mean square of pulsating pressure coefficient Represented as:

[0035]

[0036] Total sound pressure level (OASPL) is expressed as:

[0037] OASPL=20lg( / )

[0038] In step S32, the flange root strain is obtained through a ground loading test. With loading force Relationship:

[0039]

[0040] Where K is the calibration coefficient for strain and loading force;

[0041] The flange root bending moment is obtained by multiplying the applied force F by the distance S from the point of application of the force S to the flange root.

[0042]

[0043] Where S is the distance from the applied force to the wing root;

[0044] This leads to the torque M and strain. Relationship

[0045]

[0046] In wind tunnel testing, the strain values ​​of the model are collected to obtain the bending moment values, and the root mean square of the flange root strain signal is calculated. and the power spectral density of the strain signal at the wing root ;

[0047] The root mean square of the acceleration was calculated. The acceleration power spectral density was obtained by FFT transformation. ;

[0048] Root Mean Square Acceleration Represented as:

[0049]

[0050] in, This represents the change in acceleration value.

[0051] S33. Based on the parameters obtained in step S31, the pulsating pressure power spectral density is calculated using the periodogram method, and the pulsating pressure derived parameters are further obtained by solving the problem.

[0052] In step S33, the collected pulsating pressure The N data points are used as a sequence with finite energy, where N is the number of Fourier transform points, for pulsating pressure. Perform a discrete Fourier transform to obtain the discrete pulsating pressure. ;

[0053] Discretized pulsating pressure Represented as:

[0054]

[0055] in, It is a natural constant. The imaginary unit, For indexing, Angular frequency;

[0056] Discretized pulsating pressure Square the amplitude and divide by N to obtain the pulsating pressure power spectral density. The pulsating pressure one-sided power spectral density was obtained by processing it. ;

[0057] Pulsating pressure power spectral density Represented as:

[0058]

[0059] Pulsating pressure one-sided power spectral density Represented as:

[0060]

[0061] Combined with reference sound pressure The sound pressure spectrum was calculated. ;

[0062] Sound pressure spectrum Represented as:

[0063]

[0064] According to the formula The strain obtained by conversion Where K is the strain gauge sensitivity coefficient, U is the bridge supply voltage, and the full-bridge configuration is based on... Calculation of comprehensive strain ,in, The strain is due to the direction of airflow in the wind tunnel. For strain in the vertical airflow direction, then apply Hooke's Law. The converted stress is given, where E is the elastic modulus.

[0065] Furthermore, in S4, the pulsating pressure method is specifically as follows:

[0066] The pulsating pressure at the trailing edge of the wing was measured at pressure measurement points set at the half-span and the local chord length of the wing. and root mean square of pulsating pressure coefficient When flutter occurs, and Rapid changes will and The angle of attack corresponding to the transition point is defined as the buffeting initiation angle of attack. ;

[0067] The wingtip acceleration method is as follows:

[0068] An accelerometer is installed at the maximum thickness of a predetermined half-wing span section. When flutter occurs, the acceleration at the wingtip is measured. Rapid change will accelerate The angle of attack corresponding to the transition point is defined as the buffeting initiation angle of attack. ;

[0069] The flange root bending moment method is as follows:

[0070] The wing root bending moment method involves attaching resistance strain gauges at the wing root and measuring the root mean square value of an electrical signal proportional to the wing root bending moment using a testing system. (mV), The angle of attack corresponding to the rapidly increasing turning point (the intersection of two tangents) is defined as the buffeting initiation angle of attack. ;

[0071] The inflection point method for steady aerodynamic curves is as follows:

[0072] Using lift and its angle of attack Curve and axial force and their angle of attack The changing pattern of the curve will curves and The angle of attack corresponding to the first inflection point on the curve is defined as the buffeting initiation angle of attack. .

[0073] Furthermore, in S5, according to the formula = / ( b) Calculate the chatter coefficient ,in, The root mean square of the bending moment-strain signal. For reference area, b is half wingspan;

[0074] According to standards, shake is classified into levels as follows: Mild. =0.004, moderate =0.008, Severe =0.016.

[0075] Technical Solution 2: A low-speed wind tunnel buffeting test device with multi-source data fusion, used to perform a low-speed wind tunnel buffeting test method with multi-source data fusion as described in Technical Solution 1, including a data acquisition module, a data preprocessing module, a multi-source fusion calculation module, and a control and output module;

[0076] The data acquisition module includes a balance, strain gauges, a pulsating pressure sensor, and an accelerometer. The pulsating pressure sensor is arranged on the wing, the strain gauges are arranged at the wing root, the accelerometers are arranged at the wingtip, and the balance is arranged at the wing support position. The data acquisition module is connected to the data preprocessing module.

[0077] The data preprocessing module is connected to the multi-source fusion computing module, and the multi-source fusion computing module is connected to the control and output module.

[0078] The beneficial effects of this invention are as follows: 1) This invention achieves multi-source full-domain fusion: breaking through the limitations of a single data source, it realizes multi-dimensional collaborative analysis of flow field, load, and structure, resulting in a more complete characterization; 2) This invention has precise spatiotemporal registration: unifying the clock and spatial reference, solving the problem of multi-sensor asynchrony, and significantly improving the accuracy of coupled analysis; 3) This invention has strong anti-interference robustness: cross-verification of multi-source signals effectively suppresses noise and interference, making the evaluation results more reliable; 4) This invention achieves a high degree of automation: automatic processing, evaluation, and output of continuous attitude data significantly improves efficiency and reduces human error; 5) This invention has strong engineering applicability: outputting standardized data and reports directly supports aerodynamic design, structural optimization, and airworthiness certification; 6) This invention has excellent compatibility: it can be seamlessly integrated into existing low-speed wind tunnel measurement and control and acquisition systems without large-scale hardware modifications. Attached Figure Description

[0079] The accompanying drawings, which are included to provide a further understanding of the invention and form part of this invention, illustrate exemplary embodiments of the invention and are used to explain the invention, but do not constitute an undue limitation of the invention. In the drawings:

[0080] Figure 1 A flowchart illustrating a low-speed wind tunnel buffeting test method based on multi-source data fusion;

[0081] Figure 2 A schematic diagram of a low-speed wind tunnel buffeting test method based on multi-source data fusion;

[0082] Figure 3 This is a schematic diagram of the data acquisition module.

[0083] Reference numerals in the attached figures: 1. Data acquisition module; 2. Data preprocessing module; 3. Multi-source fusion calculation module; 4. Control and output module; 5. Balance; 6. Strain gauge; 7. Pulsating pressure sensor; 8. Accelerometer. Detailed Implementation

[0084] To make the technical solutions and advantages of the embodiments of the present invention clearer, the exemplary embodiments of the present invention will be further described in detail below with reference to the accompanying drawings. Obviously, the described embodiments are only a part of the embodiments of the present invention, and not an exhaustive list of all embodiments. It should be noted that, unless otherwise specified, the embodiments and features in the embodiments of the present invention can be combined with each other.

[0085] Example 1: Reference Figures 1-3 This embodiment details a low-speed wind tunnel buffeting test method based on multi-source data fusion, specifically including the following steps:

[0086] S1. Multi-source data is collected synchronously through the data acquisition module to obtain sensor data and environmental parameters of the test model;

[0087] S2. Zero-point calibration is performed on the sensor data of the test model under windless load conditions through the data preprocessing module. Signal conversion is performed on the obtained zero-point data and the sensor data of the test model to obtain the converted data.

[0088] S3. Based on the parameters obtained in steps S1 and S2, multi-dimensional parameter collaborative calculation is performed through the multi-source fusion calculation module to calculate the pulsating pressure parameters, structural response parameters, power spectral density and derived parameters.

[0089] S4. Combining the parameters obtained in steps S1-S3, determine the initial angle of attack for fluttering using the pulsating pressure method, wingtip acceleration method, wing root bending moment method, or steady aerodynamic curve inflection point method.

[0090] S5. Based on the determined initial angle of attack of the buffeting, calculate the buffeting coefficient, determine the buffeting level, automatically associate the attitude data of continuous angle of attack / sideslip angle, generate the buffeting characteristic curve, and complete the data output according to the standard format of aviation test.

[0091] Furthermore, in step S1, a pulsating pressure sensor array is arranged on the surface of the test model wing, strain gauges are attached to key stress locations at the wing root, accelerometers are arranged at the wingtip, and a force balance is installed at the wing support location. This completes the arrangement of the data acquisition module, which synchronously acquires sensor data from the test model, i.e., the pulsating pressure signal on the surface of the test model. Voltage signal of the strain gauge at the wing root wingtip accelerometer signal And simultaneously collect wind tunnel velocity and pressure Incoming static pressure Angle of attack Sideslip angle Speed ​​of sound Atmospheric pressure Reference sound pressure Environmental parameters such as temperature T are used to ensure that all signals are triggered by the same clock and sampled using the same time base.

[0092] Furthermore, in step S2, baseline pressure data under windless conditions is collected. Combined with incoming static pressure Zero-point data were calculated. ;

[0093] Zero point data Represented as

[0094]

[0095] Among them, if the differential pressure sensor is the static pressure of the incoming flow... Take 0, To perform an averaging operation;

[0096] Zero-point data is converted through pulsed pressure conversion. and the pulsating pressure signal on the surface of the test model The data is converted into a pressure value, which is the converted pressure value obtained from the pulsating pressure sensor during the experiment.

[0097] Converted pressure value Represented as:

[0098]

[0099] Where a is the coefficient of the pulsating pressure sensor;

[0100] The DC offset and noise are removed from the time-domain signal to obtain the converted acceleration signal.

[0101] Specifically, signal conversion involves converting voltage / charge electrical signals into physical quantities such as pressure, acceleration, strain, force, and sound pressure according to the sensor calibration coefficients, removing DC components, eliminating abnormal data, and using adaptive filtering to reduce noise and improve the signal-to-noise ratio.

[0102] Furthermore, step S3 includes the following steps:

[0103] S31. Calculate the total sound pressure level based on the sensor data of the test model and the converted data;

[0104] In step S31, the local static pressure is obtained by averaging the converted pressure values. Take the converted pressure value and local static pressure The difference in value is obtained from the pulsating pressure. Further solving yields the root mean square of the pulsating pressure. Combined with wind tunnel speed pressure Solving for the root mean square of the pulsating pressure coefficient yields the solution. Combined with reference sound pressure Solving for the total sound pressure level (OASPL), we obtain the total sound pressure level (OASPL).

[0105] Local static pressure Represented as:

[0106] =mean( )

[0107] Pulsating pressure Represented as:

[0108]

[0109] Root mean square of pulsating pressure Represented as:

[0110]

[0111] Root mean square of pulsating pressure coefficient Represented as:

[0112]

[0113] Total sound pressure level (OASPL) is expressed as:

[0114] OASPL=20lg( / )

[0115] Among them, the sound pressure ratio is converted into the total sound pressure level by using 20lg;

[0116] S32. Based on the sensor data of the test model and the converted data, calculate the relevant parameters of the wing root bending moment and the wing tip acceleration.

[0117] In step S32, the flange root strain is obtained through a ground loading test. With loading force Relationship:

[0118]

[0119] Where K is the calibration coefficient for strain and loading force;

[0120] The flange root bending moment is obtained by multiplying the applied force F by the distance S from the point of application of the force S to the flange root.

[0121]

[0122] Where S is the distance from the applied force to the wing root;

[0123] This leads to the torque M and strain. Relationship

[0124]

[0125] In wind tunnel testing, the strain values ​​of the model are collected to obtain the bending moment values, and the root mean square of the flange root strain signal is calculated. and the power spectral density of the strain signal at the wing root ;

[0126] The root mean square of the acceleration was calculated. The acceleration power spectral density was obtained by FFT transformation. ;

[0127] Root Mean Square Acceleration Represented as:

[0128]

[0129] in, This represents the change in acceleration value.

[0130] S33. Based on the parameters obtained in step S31, the pulsating pressure power spectral density is calculated using the periodogram method, and the pulsating pressure derived parameters are further obtained by solving the problem.

[0131] In step S33, the collected pulsating pressure The N data points are used as a sequence with finite energy, where N is the number of Fourier transform points, for pulsating pressure. Perform a discrete Fourier transform to obtain the discrete pulsating pressure. ;

[0132] Discretized pulsating pressure Represented as:

[0133]

[0134] in, It is a natural constant. The imaginary unit, For indexing, Angular frequency;

[0135] Discretized pulsating pressure Square the amplitude and divide by N to obtain the pulsating pressure power spectral density. The pulsating pressure one-sided power spectral density was obtained by processing it. ;

[0136] Pulsating pressure power spectral density Represented as:

[0137]

[0138] Power spectral density Symmetric about the Y-axis, the frequency range is (- This range is for mathematical convenience; negative frequencies do not actually exist. (The text also mentions the single-sided power spectral density of pulsating pressure.) Represented as:

[0139]

[0140] Combined with reference sound pressure The sound pressure spectrum was calculated. ;

[0141] Sound pressure spectrum Represented as:

[0142]

[0143] According to the formula The strain obtained by conversion Where K is the strain gauge sensitivity coefficient, U is the bridge supply voltage, and the full-bridge configuration is based on... Calculation of comprehensive strain ,in, The strain is due to the direction of airflow in the wind tunnel. For strain in the vertical airflow direction, then apply Hooke's Law. The converted stress is given, where E is the elastic modulus.

[0144] Furthermore, in S4, the pulsating pressure method is specifically as follows:

[0145] Pressure measurement points were set at 75%–95% of the wing's half-span and 90%–95% of the local chord length to measure the pulsating pressure at the wing's trailing edge. and root mean square of pulsating pressure coefficient When flutter occurs, and Rapid changes will and The angle of attack corresponding to the transition point is defined as the buffeting initiation angle of attack. ;

[0146] The wingtip acceleration method is as follows:

[0147] The accelerometer was installed at the maximum thickness of the 90%–95% half-wing span section. When flutter occurred, the acceleration at the wingtip was measured. Rapid change will accelerate The angle of attack corresponding to the transition point is defined as the buffeting initiation angle of attack. ;

[0148] The flange root bending moment method is as follows:

[0149] The wing root bending moment method involves attaching resistance strain gauges at the wing root and measuring the root mean square value of an electrical signal proportional to the wing root bending moment using a testing system. (mV), The angle of attack corresponding to the rapidly increasing turning point (the intersection of two tangents) is defined as the buffeting initiation angle of attack. ;

[0150] The inflection point method for steady aerodynamic curves is as follows:

[0151] Using lift and its angle of attack Curve and axial force and their angle of attack The curve's variation pattern determines the starting point of buffeting; when airflow adheres to the wing surface... The curve is basically linear; once airflow separation occurs on the wing surface, The linear relationship of the curve is broken. The slope of the curve changes, curves and The angle of attack corresponding to the first inflection point on the curve is defined as the buffeting initiation angle of attack. .

[0152] Furthermore, in S5, according to the formula = / ( b) Calculate the chatter coefficient ,in, The root mean square of the bending moment-strain signal. For reference area, b is half wingspan;

[0153] According to standards, shake is classified into levels as follows: Mild. =0.004, moderate =0.008, Severe =0.016.

[0154] Specifically, this embodiment includes the following steps:

[0155] 1) Multi-source data acquisition: A 16-channel pulsating pressure sensor is installed on the wing, two full-bridge strain gauges are installed at the wing root, one accelerometer is installed at the wingtip, and one force balance is installed at the wing support; angle of attack range 0°. 20°, step size 2°, synchronously acquire pressure, acceleration, strain, force, velocity pressure, angle of attack, and static pressure signals;

[0156] 2) Data preprocessing: Acquire windless reference signals, complete zero-point calibration and DC offset subtraction; convert sensor electrical signals into pressure, acceleration, and strain physical quantities; perform filtering and noise reduction and outlier removal;

[0157] 3) Multi-dimensional parameter calculation: Calculate the root mean square of pulsating pressure, pulsating pressure coefficient, total sound pressure level, power spectral density, root mean square of acceleration, and root mean square of flange root bending moment at each angle of attack;

[0158] 4) Spatiotemporal registration and fusion: Time synchronization and spatial coordinate registration are completed, and after feature-level and decision-level fusion, the starting angle of attack for buffeting is identified as 12°;

[0159] 5) Shake level assessment and data output: Calculate the shake coefficient C b =0.005, which is judged as mild buffeting according to fighter jet standards; automatically generate characteristic curves of pulsating pressure coefficient-angle of attack, power spectral density-frequency, and buffeting coefficient-angle of attack, and output test reports in aviation standard format.

[0160] Example 2: Reference Figures 1-3 This embodiment describes a low-speed wind tunnel buffeting test device based on multi-source data fusion, used to perform a low-speed wind tunnel buffeting test method based on multi-source data fusion as described in Embodiment 1. The device includes a data acquisition module 1, a data preprocessing module 2, a multi-source fusion calculation module 3, and a control and output module 4.

[0161] The data acquisition module 1 includes a balance 5, a strain gauge 6, a pulsating pressure sensor 7, and an accelerometer 8. The pulsating pressure sensor 7 is arranged on the wing, the strain gauge 6 is arranged at the wing root, the accelerometer 8 is arranged at the wingtip, and the balance 5 is arranged at the wing support position. The data acquisition module 1 is connected to the data preprocessing module 2.

[0162] The data preprocessing module 2 is connected to the multi-source fusion computing module 3, and the multi-source fusion computing module 3 is connected to the control and output module 4.

[0163] Specifically, data acquisition module 1 has 16... The 64-channel synchronous acquisition is used to synchronously acquire pulsating pressure, acceleration, strain, lift, axial force, wind tunnel environment, and attitude angle signals. It includes a sensor array, signal conditioning unit, and synchronous acquisition card. The pulsating pressure sensor 7 uses 16 channels, and the strain gauge 6 is a 2-channel full-bridge strain gauge 6.

[0164] Data preprocessing module 2 is used for zero-point calibration, DC offset removal, filtering and noise reduction, electrical signal-to-physical quantity conversion, and outlier removal;

[0165] The multi-source fusion computing module 3 is used for multi-dimensional parameter calculation, spatiotemporal registration, feature-level fusion, decision-level fusion, chattering coefficient calculation and level evaluation;

[0166] The control and output module 4 is used for automatic control of the test process, continuous attitude data association, characteristic curve generation, standardized data output and visualization display. It can be directly connected to the existing large-scale low-speed wind tunnel measurement and control system to realize the full-process automation and standardized operation of the vibration test.

[0167] Although the invention has been described with reference to a limited number of embodiments, those skilled in the art will understand from the foregoing description that other embodiments are conceivable within the scope of the invention described herein. Furthermore, it should be noted that the language used in this specification has been chosen primarily for readability and instructional purposes, and not for the purpose of interpreting or limiting the subject matter of the invention. Therefore, many modifications and variations will be apparent to those skilled in the art without departing from the scope and spirit of the appended claims. The disclosure of the invention is illustrative and not restrictive, and the scope of the invention is defined by the appended claims.

Claims

1. A low-speed wind tunnel buffeting test method using multi-source data fusion, characterized in that, Includes the following steps: S1. Multi-source data is collected synchronously through the data acquisition module to obtain sensor data and environmental parameters of the test model; S2. Zero-point calibration is performed on the sensor data of the test model under windless load conditions through the data preprocessing module. Signal conversion is performed on the obtained zero-point data and the sensor data of the test model to obtain the converted data. S3. Based on the parameters obtained in steps S1 and S2, multi-dimensional parameter collaborative calculation is performed through the multi-source fusion calculation module to calculate the pulsating pressure parameters, structural response parameters, power spectral density and derived parameters. S4. Combining the parameters obtained in steps S1-S3, determine the initial angle of attack for fluttering using the pulsating pressure method, wingtip acceleration method, wing root bending moment method, or steady aerodynamic curve inflection point method. S5. Based on the determined initial angle of attack of the buffeting, calculate the buffeting coefficient, determine the buffeting level, automatically associate the attitude data of continuous angle of attack / sideslip angle, generate the buffeting characteristic curve, and complete the data output according to the standard format of aviation test.

2. The low-speed wind tunnel buffeting test method based on multi-source data fusion according to claim 1, characterized in that, In step S1, a pulsating pressure sensor array is arranged on the surface of the test model wing, strain gauges are attached to key stress locations at the wing root, accelerometers are arranged at the wingtip, and a force balance is installed at the wing support location. This completes the data acquisition module setup and synchronously acquires sensor data from the test model, i.e., the pulsating pressure signal on the test model surface. Voltage signal of the strain gauge at the wing root wingtip accelerometer signal And simultaneously collect wind tunnel velocity and pressure Incoming static pressure Angle of attack Sideslip angle Speed ​​of sound Atmospheric pressure Reference sound pressure Including environmental parameters such as temperature T, ensure that all signals are triggered by the same clock and sampled using the same time base.

3. The low-speed wind tunnel buffeting test method based on multi-source data fusion according to claim 2, characterized in that, In step S2, baseline pressure data under windless conditions is collected. Combined with incoming static pressure Zero-point data were calculated. ; Zero point data Represented as ； in, To perform an averaging operation; Zero-point data is converted through pulsed pressure conversion. and the pulsating pressure signal on the surface of the test model The data is converted into a pressure value, which is the converted pressure value obtained from the pulsating pressure sensor during the experiment. Converted pressure value Represented as: ； Where a is the coefficient of the pulsating pressure sensor; The DC offset and noise are removed from the time-domain signal to obtain the converted acceleration signal.

4. The low-speed wind tunnel buffeting test method based on multi-source data fusion according to claim 3, characterized in that, S3 includes the following steps: S31. Calculate the total sound pressure level based on the sensor data of the test model and the converted data; In step S31, the local static pressure is obtained by averaging the converted pressure values. Take the converted pressure value and local static pressure The difference in value is obtained from the pulsating pressure. Further solving yields the root mean square of the pulsating pressure. Combined with wind tunnel speed pressure Solving for the root mean square of the pulsating pressure coefficient yields the solution. Combined with reference sound pressure Solving for the total sound pressure level (OASPL), we obtain the total sound pressure level (OASPL). Local static pressure Represented as: =mean( ); Pulsating pressure Represented as: ； Root mean square of pulsating pressure Represented as: ； Root mean square of pulsating pressure coefficient Represented as: ； Total sound pressure level (OASPL) is expressed as: OASPL=20lg( / ); Among them, the sound pressure ratio is converted into the total sound pressure level by using 20lg; S32. Based on the sensor data of the test model and the converted data, calculate the relevant parameters of the wing root bending moment and the wing tip acceleration. In step S32, the flange root strain is obtained through a ground loading test. With loading force Relationship: ； Where K is the calibration coefficient for strain and loading force; The flange root bending moment is obtained by multiplying the applied force F by the distance S from the point of application of the force S to the flange root. ； Where S is the distance from the applied force to the wing root; This leads to the torque M and strain. Relationship ； In wind tunnel testing, the strain values ​​of the model are collected to obtain the bending moment values, and the root mean square of the flange root strain signal is calculated. and the power spectral density of the strain signal at the wing root ; The root mean square of the acceleration was calculated. The acceleration power spectral density was obtained by FFT transformation. ; Root Mean Square Acceleration Represented as: ； in, This represents the change in acceleration value. S33. Based on the parameters obtained in step S31, the pulsating pressure power spectral density is calculated using the periodogram method, and the pulsating pressure derived parameters are further solved. In step S33, the collected pulsating pressure The N data points are used as a sequence with finite energy, where N is the number of Fourier transform points, for pulsating pressure. Perform a discrete Fourier transform to obtain the discrete pulsating pressure. ; Discretized pulsating pressure Represented as: ； in, It is a natural constant. The imaginary unit, For indexing, Angular frequency; Discretized pulsating pressure Square the amplitude and divide by N to obtain the pulsating pressure power spectral density. The pulsating pressure one-sided power spectral density was obtained by processing it. ; Pulsating pressure power spectral density Represented as: ； Pulsating pressure one-sided power spectral density Represented as: ； Combined with reference sound pressure The sound pressure spectrum was calculated. ; Sound pressure spectrum Represented as: ； According to the formula The strain obtained by conversion Where K is the strain gauge sensitivity coefficient, U is the bridge supply voltage, and the full-bridge configuration is based on... Calculation of comprehensive strain ,in, The strain is due to the direction of airflow in the wind tunnel. For strain in the vertical airflow direction, then apply Hooke's Law. The converted stress is given, where E is the elastic modulus.

5. The low-speed wind tunnel buffeting test method based on multi-source data fusion according to claim 4, characterized in that, In S4, the pulsating pressure method is specifically as follows: The pulsating pressure at the trailing edge of the wing was measured at pressure measurement points set at the half-span and the local chord length of the wing. and root mean square of pulsating pressure coefficient When flutter occurs, and Rapid changes will and The angle of attack corresponding to the transition point is defined as the buffeting initiation angle of attack. ; The wingtip acceleration method is as follows: An accelerometer is installed at the maximum thickness of a predetermined half-wing span section. When flutter occurs, the acceleration at the wingtip is measured. Rapid change will accelerate The angle of attack corresponding to the transition point is defined as the buffeting initiation angle of attack. ; The flange root bending moment method is as follows: The wing root bending moment method involves attaching resistance strain gauges at the wing root and measuring the root mean square value of an electrical signal proportional to the wing root bending moment using a testing system. , The angle of attack corresponding to the rapidly increasing transition point is defined as the buffeting initiation angle of attack. ; The inflection point method for steady aerodynamic curves is as follows: Using lift and its angle of attack Curve and axial force and their angle of attack The changing pattern of the curve will curves and The angle of attack corresponding to the first inflection point on the curve is defined as the buffeting initiation angle of attack. .

6. The low-speed wind tunnel buffeting test method based on multi-source data fusion according to claim 5, characterized in that, In S5, according to the formula = / ( b) Calculate the chatter coefficient ,in, The root mean square of the bending moment-strain signal. For reference area, b is half wingspan; According to standards, shake is classified into levels as follows: Mild. =0.004, moderate =0.008, Severe =0.

016.

7. A low-speed wind tunnel vibration testing device with multi-source data fusion, characterized in that, The method for performing a low-speed wind tunnel vibration test by multi-source data fusion as described in any one of claims 1-6 includes a data acquisition module (1), a data preprocessing module (2), a multi-source fusion calculation module (3), and a control and output module (4). The data acquisition module (1) includes a balance (5), a strain gauge (6), a pulsating pressure sensor (7) and an accelerometer (8). The pulsating pressure sensor (7) is arranged on the wing, the strain gauge (6) is arranged at the wing root, the accelerometer (8) is arranged at the wingtip, and the balance (5) is arranged at the wing support position. The data acquisition module (1) is connected to the data preprocessing module (2). The data preprocessing module (2) is connected to the multi-source fusion computing module (3), and the multi-source fusion computing module (3) is connected to the control and output module (4).

Images