A real-time control method for multi-dimensional vibration of wind tunnel model support system

Abstract

The present invention belongs to the field of active control of aircraft model vibration, and discloses a method for real-time control of multi-dimensional vibration of a wind tunnel model support system. The method of the present invention first designs a multi-dimensional vibration monitoring layout and a multi-dimensional vibration suppression layout of a wind tunnel model support system, synchronously collects aerodynamic load and inertial load design, and designs a multi-actuator collaborative control matrix; on this basis, based on the Hilbert-Huang transform filtering load signal to the target frequency domain, it is equivalent to a dynamic load vector on the vibration suppression actuation surface, and a two-dimensional PID controller is constructed based on the principle of torque balance to achieve effective control of multi-dimensional vibration. Real-time and reliable control of multi-dimensional vibration in a complex flow field environment is achieved. The method of the present invention can ensure the reliable implementation of wind tunnel tests, expand the test scope of wind tunnel tests, and reduce the production cost of wind tunnel tests.

Description

Technical Field

[0001] The invention belongs to the field of active vibration control of aircraft models and relates to a real-time control method for multi-dimensional vibration of a wind tunnel model support system. Background Art

[0002] In the process of developing high-performance aircraft, wind tunnel model tests can be used to test its key flight characteristics and aerodynamic data, which can effectively reduce the cost of aircraft development and improve aerodynamic performance. Due to the small interference to the flow field around the aircraft, a cantilever tail support structure is often used to fix the aircraft wind tunnel model, and the wind tunnel model aerodynamic force is measured by an embedded aerodynamic balance. However, during the test, the cantilever structure has the characteristics of weak stiffness and low damping, and is prone to violent vibration under the action of aerodynamic loads, which not only interferes with the difficulty of measuring aerodynamic data, but also easily causes damage to the aerodynamic balance and even the tail support rod in severe cases. Therefore, it is necessary to actively control the vibration of the wind tunnel model support system caused by aerodynamic loads in real time. At present, active vibration control for wind tunnel tests under fixed angle of attack has achieved good results, but the aerodynamic load in high dynamic wind tunnel tests shows multi-directional and transient characteristics. In order to ensure the reliable measurement of aerodynamic force, extremely high requirements are put forward for the measurement, layout and control methods of the active vibration control system. Therefore, it is of great significance to study the real-time control method of wind tunnel model-support system vibration for multi-directional aerodynamic loads.

[0003] Liu Yu and others from the Shenyang Aerodynamics Research Institute of the Aviation Industry Corporation of China proposed an active vibration suppression method in the patent No. 202210922741.X published in 2022, "Large aspect ratio model longitudinal vibration front and rear shock absorber coordination control method and system", which uses an accelerometer as a feedback signal and drives two sets of piezoelectric ceramic shock absorbers to control the vibration of different positions of the "model-balance-strut system" based on the principle of torque balance. The device can effectively control the multi-order vibrations generated at different positions of the aircraft model during the test, but this method is designed for fixed angle of attack tests and is difficult to cope with the severe vibrations caused by multi-directional and transient aerodynamic loads during high dynamic tests.

[0004] At present, the active vibration suppression method of the model based on piezoelectric ceramic actuators is mainly designed for fixed angle of attack test conditions. However, in dynamic wind tunnel tests, the aerodynamic loads on the dynamic wind tunnel model under complex flow fields are subjected to multiple load directions and large instantaneous changes, which puts higher requirements on the high real-time performance of the active control method. The existing methods cannot meet the needs of reliable wind tunnel tests. Therefore, it is an effective method to control the vibration of the wind tunnel model support system caused by multi-directional transient aerodynamic loads by building a measurement layout for aerodynamic loads and inertial loads, designing a two-dimensional vibration suppression layout and solving the collaborative control matrix, solving the dynamic load vector in the target frequency domain at the vibration suppression actuation surface, and realizing real-time control of aerodynamic loads based on the principle of torque balance. Summary of the invention

[0005] The main technical problem solved by the present invention is to overcome the deficiencies of the prior art, and a real-time control method for multi-dimensional vibration of a wind tunnel model support system is proposed. The method first designs the vibration monitoring layout of the wind tunnel model support system, and synchronously collects the aerodynamic load and inertial load; further, a multi-dimensional vibration suppression layout is designed, and the multi-actuator collaborative control matrix is ​​designed taking into account the stiffness differences in pitch and yaw dimensions; on this basis, the load signal is filtered based on the Hilbert-Huang transform to suppress noise interference, and the filtered load signal is equivalent to the dynamic load vector on the vibration suppression actuation surface, and a two-dimensional PID controller is constructed based on the principle of torque balance to achieve effective control of multi-dimensional vibration. The active control effect is good for the vibration caused by multi-dimensional transient aerodynamic loads during the wind tunnel dynamic test, which can greatly reduce the production cost of the wind tunnel test.

[0006] The technical solution of the present invention:

[0007] A real-time control method for multi-dimensional vibration of a wind tunnel model support system is proposed. The method synchronously collects the aerodynamic load and inertial load of the wind tunnel model support system during vibration, constructs a multi-actuator collaborative control matrix considering the stiffness difference between pitch and yaw dimensions, and outputs reverse resistance based on the principle of moment balance to achieve effective vibration control. The steps are as follows:

[0008] (1) Build a vibration suppression system for the wind tunnel model support system under multi-directional dynamic aerodynamic loads, and simultaneously collect the inertial load and aerodynamic load data of the wind tunnel model support system;

[0009] (2) In response to the vibration of the wind tunnel model support system caused by multi-directional dynamic aerodynamic loads, a vibration suppressor layout with pitch and yaw dimensions was designed. Four piezoelectric ceramic actuators were evenly distributed along the circumference of the OXY plane to form an "X"-shaped layout. A collaborative control matrix was designed based on the differences in pitch and yaw dimensions of the wind tunnel model.

[0010] (3) filtering the pneumatic load and inertial load data collected in step (1); and transmitting the dynamic load data including the pneumatic load and the inertial load in the target frequency domain to the vibration suppression actuating surface, which is equivalent to the dynamic load vector on the vibration suppression actuating surface;

[0011] (4) Based on the principle of torque balance, PID controllers are constructed in the pitch and yaw dimensions respectively. Taking the dynamic load vector constructed in step (3) as the target, the vibration suppression forces to be output in the pitch and yaw dimensions are calculated respectively. Then, based on the coordinated control matrix of the multi-piezoelectric ceramic actuator constructed in step (2) and the amplitude of the dynamic load vector, the input control voltage of each piezoelectric ceramic actuator is solved. Finally, each piezoelectric ceramic actuator is driven to output the vibration suppression force to realize the control of the multi-dimensional vibration of the wind tunnel model support system.

[0012] Furthermore, in step (1), a vibration suppression system for a wind tunnel model support system under multi-directional dynamic aerodynamic loads is constructed, which includes an aircraft wind tunnel model, an acceleration sensor, a wind tunnel balance, a tail support rod, a vibration suppressor and a fixing mechanism; two acceleration sensors are respectively arranged in the pitch and yaw dimensions at the center of mass of the aircraft wind tunnel model, and synchronously collect the inertial load and aerodynamic load data of the wind tunnel model support system with the wind tunnel balance, thereby realizing high-precision and high-dynamic collection of multi-directional dynamic aerodynamic loads.

[0013] Furthermore, the specific implementation process of step (2) is as follows:

[0014] During the vibration suppression process, four piezoelectric ceramic actuators work together to control the vibration caused by multi-directional dynamic aerodynamic loads in the pitch and yaw directions. Due to the differences in dynamic stiffness in pitch and yaw dimensions between different wind tunnel models, the main control angles of the piezoelectric ceramic actuators under different vibration states are defined for different wind tunnel models and their frequency-varying processes. , and define an orthogonal base coordinate system OXY with the center of the actuating section of the piezoelectric ceramic actuator as the origin, and the yaw and pitch dimensions as the X and Y axes, and a non-orthogonal base coordinate system OX1Y1 with the center of the actuating section of the same piezoelectric ceramic actuator as the origin, and the X1 and Y1 axes determined by the main control direction. The conversion relationship between the two is as follows:

[0015] (1)

[0016] in, is the coordinate of the external load force in the orthogonal base coordinate system OXY, is the coordinate in the non-orthogonal base coordinate system OX1Y1; thus, the aerodynamic load of the orthogonal base coordinate system is converted to the non-orthogonal base coordinate system. When the piezoelectric ceramic actuator outputs the response vibration suppression force according to the converted external load, it is equivalent to the pitch and yaw dimensions of the wind tunnel model support system being in a state of equal stiffness. The control quantity of the piezoelectric ceramic actuator with an "X"-shaped inclined 45° layout is directly solved and output without further processing;

[0017] The coordinated control matrix of multiple piezoelectric ceramic actuators is derived as follows: They represent the control quantity of the piezoelectric ceramic actuator respectively;

[0018] (2).

[0019] Furthermore, the specific implementation process of step (3) is as follows:

[0020] (3.1) Based on the vibration suppression system built in step (1), four piezoelectric ceramic actuators are assembled, the power cord is connected to input the control voltage, and the preload components are matched by wedge surfaces to complete the assembly of the vibration suppression structure;

[0021] (3.2) Four piezoelectric ceramic actuators are used to output a 0~x Hz sinusoidal sweep signal, which drives the wind tunnel model support system to vibrate, and the time domain response acceleration of the wind tunnel model support system is collected through an acceleration sensor; based on the collected time domain response acceleration data processed by fast Fourier transform, a power spectrum density diagram of the time domain response acceleration is drawn, and the first extreme value of the power spectrum density is found from 0 Hz, and its corresponding frequency is the first-order natural frequency of the wind tunnel model support system;

[0022] (3.3) A filter is constructed based on the Hilbert-Huang transform to process the collected aerodynamic load and inertial load data and remove the noise signals outside the frequency domain of the first-order natural frequency:

[0023] First, based on empirical mode decomposition, the upper and lower envelopes of the collected pneumatic load and inertial load data are taken, and the average values ​​of the corresponding points of the upper and lower envelopes are connected into a line, which is the average line; the average line is subtracted from the original load data to obtain a new signal, which is an intrinsic mode function IMF. Repeat this process to decompose the pneumatic load and inertial load signals into several IMFs, and each IMF corresponds to a frequency signal;

[0024] Then, each IMF is Hilbert transformed as follows, and the instantaneous frequency, amplitude and phase of each order of the mixed signal are obtained after the transformation, and then the time-frequency spectrum of the frequency signal is constructed;

[0025] (3)

[0026] in, represents the time domain time, represents the real signal before transformation, represents the result of Hilbert transform, Express Perform a Hilbert transform, represents the Cauchy principal components, represents the integral variable, Represents the original signal exist The value of the moment, Indicates the current time point And historical time point The time difference, Represents a and The analytical signal composed of represents the instantaneous amplitude of the signal, represents the instantaneous phase of the signal, represents the instantaneous frequency of the signal, express about The derivative of express about The derivative of

[0027] Finally, the maximum frequencies corresponding to the instantaneous amplitudes in the frequency signals of each order are obtained after the empirical mode decomposition in the pitch and yaw dimensions respectively as the first-order natural frequencies of the wind tunnel model support system in the pitch and yaw dimensions. At the same time, the signals other than the first-order natural frequencies of the vibration in the pitch and yaw dimensions are eliminated. Finally, a signal filter based on the Hilbert-Huang transform is constructed to eliminate the frequency domain noise interference signals other than the first-order natural frequencies of the wind tunnel model support system.

[0028] (3.4) The filtered pneumatic load and inertial load data are converted to the vibration suppression surface according to the following formula, and are jointly equivalent to a dynamic load vector;

[0029] (4)

[0030] in, It represents the magnitude of the dynamic load vector on the vibration suppression surface. represents the equivalent mass of the cantilever support, wind tunnel balance and wind tunnel model before the vibration suppression surface. Indicates the yaw acceleration and pitch acceleration The total acceleration , Indicates the distance from the acceleration sensor to the vibration suppression surface. represents the multidimensional aerodynamic force measured by the wind tunnel balance, It indicates the distance from the force measuring center of the wind tunnel balance to the vibration suppression surface.

[0031] The beneficial effect of the present invention is to provide a real-time control method for multi-dimensional vibration of a wind tunnel model support system. First, a multi-dimensional vibration monitoring layout and a multi-dimensional vibration suppression layout of the wind tunnel model support system are designed, the pneumatic load and inertial load design are collected synchronously, and a multi-actuator collaborative control matrix is ​​designed; on this basis, the load signal is filtered to the target frequency domain based on the Hilbert-Huang transform, which is equivalent to a dynamic load vector on the vibration suppression actuation surface, and a two-dimensional PID controller is constructed based on the principle of torque balance to achieve effective control of multi-dimensional vibration. Real-time and reliable control of multi-dimensional vibration in a complex flow field environment is achieved. This method can ensure the reliable implementation of wind tunnel tests, expand the test scope of wind tunnel tests, and reduce the production cost of wind tunnel tests. BRIEF DESCRIPTION OF THE DRAWINGS

[0032] Figure 1 The figure is a flowchart of the method of the present invention.

[0033] Figure 2 This is a layout diagram of the vibration suppression system of the wind tunnel model support system of the present invention.

[0034] Figure 3 It is the layout diagram of the vibration suppressor of the present invention.

[0035] Figure 4 It is the load conversion coordinate diagram of the present invention.

[0036] Figure 5 This is a diagram of the filtering results of the acceleration signal in the yaw dimension of an inertial load according to the present invention.

[0037] Figure 6 This is another graph of the filtering results of the pitch dimension acceleration signal of an inertial load according to the present invention.

[0038] Figure 7 This is a diagram showing the effect of yaw dimension vibration control according to the present invention.

[0039] Figure 8 This is another pitch dimension vibration control effect diagram of the present invention.

[0040] In the figure: 1 aircraft wind tunnel model; 2 pitch dimension acceleration sensor; 3 yaw dimension acceleration sensor; 4 wind tunnel balance; 5 tail support rod; 6 vibration suppressor; 7 fixing mechanism. DETAILED DESCRIPTION

[0041] The specific implementation of the present invention is described in detail below in conjunction with specific embodiments and drawings.

[0042] This embodiment takes a certain large aspect ratio aircraft model as an example, and is described as follows in conjunction with the accompanying drawings:

[0043] (1) Construct a vibration suppression system for the wind tunnel model support system under multi-directional transient aerodynamic loads, including a large aspect ratio aircraft model, pitch and yaw acceleration sensors, a wind tunnel balance, a tail strut, a vibration suppressor and a fixing mechanism. Two acceleration sensors are respectively arranged in the pitch and yaw dimensions at the center of mass of the large aspect ratio aircraft model, and synchronously collect the inertial load and aerodynamic load of the wind tunnel model support system with the wind tunnel balance, realizing high-precision and high-dynamic collection of multi-directional transient aerodynamic loads.

[0044] (2) In order to solve the vibration of the wind tunnel model support system caused by multi-directional aerodynamic loads, a vibration suppressor is designed with a pitch and yaw dual-dimensional vibration suppression layout. Four piezoelectric ceramic actuators are evenly distributed along the circumference of the OXY plane to form an "X-shaped" layout. The vibration suppressor layout structure is as follows: Figure 3 shown.

[0045] During the vibration suppression process, four piezoelectric ceramic actuators work together to control the vibration caused by multi-directional dynamic aerodynamic loads in the pitch and yaw directions. Taking into account the differences in dynamic stiffness in pitch and yaw between different wind tunnel models, the main control angles of the piezoelectric ceramic actuators under different vibration states are defined for different wind tunnel models and their frequency-varying processes. , for a large aspect ratio aircraft model, its main control angle is obtained It is 48.5° in the non-resonance zone, 21° in the yaw resonance zone, and 85° in the pitch resonance zone.

[0046] The orthogonal base coordinate system OXY with the center of the actuating section of the piezoelectric ceramic actuator as the origin and the yaw and pitch dimensions as the X and Y axes, and the non-orthogonal base coordinate system OX1Y1 with the same center as the origin and the X1 and Y1 axes determined by the main control direction are defined separately. The relationship between the coordinate systems is as follows: Figure 4 The conversion relationship between the two is shown in formula (1).

[0047] (1)

[0048] in, is the coordinate of the external load force in the orthogonal base coordinate system OXY, and is the coordinate in the non-orthogonal base coordinate system OX1Y1. Therefore, the pneumatic external load of the orthogonal base coordinate system can be converted to the non-orthogonal base coordinate system. When the piezoelectric ceramic actuator outputs the response vibration suppression force according to the converted external load, it is equivalent to the pitch and yaw dimensions of the model support system being in a state of equal stiffness. The control quantity of the piezoelectric actuator with an X-shaped 45° inclined layout can be directly solved and output without further processing.

[0049] The coordinated control matrix of the multi-piezoelectric ceramic actuator is further derived as shown in formula (2): Respectively Figure 4 Control quantity of actuators No. 1 to 4.

[0050] (2)

[0051] (3) A filter is constructed based on the Hilbert-Huang transform to process the collected aerodynamic load and inertial load signals and remove the noise signals outside the frequency domain of their first-order natural frequency, thus laying the foundation for the real-time control of multi-dimensional vibration.

[0052] (3.1) Based on the two-dimensional vibration suppression structure designed in step (1), four piezoelectric ceramic actuators are assembled, and the power lines are connected to realize the input of control voltage. The preload components are matched through the wedge surface to complete the assembly of the vibration suppression structure.

[0053] (3.2) Based on the assembled vibration suppression structure, four piezoelectric ceramic actuators output a 0-60 Hz sine frequency sweep signal, which drives the wind tunnel model support system to vibrate. The acceleration sensor arranged at the front end collects the time domain response acceleration of the wind tunnel model support system. Based on the fast Fourier transform processing of the collected time domain response acceleration data, the power spectrum density diagram of the time domain response acceleration is drawn, and the first extreme value of the power spectrum density is found from 0 Hz. The corresponding frequency is the first-order natural frequency of the wind tunnel model support system.

[0054] (3.3) In order to solve the problem that the noise interference in the frequency domain outside the first-order natural frequency of the wind tunnel model support system causes the actuator to vibrate in the frequency domain near the first-order natural frequency during the vibration suppression process, the controller settles the control voltage based on the noise interference signal and prompts the vibration suppressor to output the vibration suppression force, which in turn stimulates the model support system to vibrate, the collected aerodynamic load and inertial load signals are processed based on the Hilbert-Huang transform:

[0055] First, based on empirical mode decomposition, the upper and lower envelopes of the collected pneumatic load and inertial load signals are taken, and the average values ​​of the corresponding points of the upper and lower envelopes are connected into a line. The average line is subtracted from the original load signal to obtain a new signal, which is an intrinsic mode function IMF. Repeating this process can decompose the pneumatic load and inertial load signals into several IMFs, and each IMF corresponds to a frequency signal.

[0056] Then, each IMF is Hilbert transformed according to the following formula, and the instantaneous frequency, amplitude and phase of each order of the mixed signal itself are obtained after the transformation, and then the time-frequency spectrum of the frequency signal can be constructed;

[0057] (3)

[0058] in, represents the time domain time, represents the real signal before transformation, represents the result of Hilbert transform, Express Perform a Hilbert transform, represents the Cauchy principal components, represents the integral variable, Represents the original signal exist The value of the moment, Indicates the current time point And historical time point The time difference, Represents a and The analytical signal composed of represents the instantaneous amplitude of the signal, represents the instantaneous phase of the signal, represents the instantaneous frequency of the signal, express about The derivative of express about The derivative of .

[0059] Finally, the frequencies with the largest instantaneous amplitudes in the frequency signals of each order are obtained after empirical mode decomposition in the pitch and yaw dimensions, respectively, as the first-order natural frequencies of the wind tunnel model support system in the pitch and yaw dimensions. At the same time, considering that in the wind tunnel test vibration control, the focus is on low-frequency and high-energy vibrations, and the signals outside the first-order natural frequencies of the two-dimensional vibrations are eliminated. Finally, a signal filter based on the Hilbert-Huang transform is constructed to eliminate the frequency domain noise interference signals outside the first-order natural frequencies of the wind tunnel model support system. Among them, the results of filtering the acceleration signals collected in the yaw and pitch dimensions are shown as follows: Figure 5 , Figure 6 shown.

[0060] (3.4) The filtered pneumatic load and inertial load are converted to the vibration suppression surface according to formula (4), and are jointly equivalent to a dynamic load vector, laying the foundation for real-time vibration control.

[0061] (4)

[0062] in, It represents the magnitude of the dynamic load vector on the vibration suppression surface. represents the equivalent mass of the cantilever support, wind tunnel balance and wind tunnel model before the vibration suppression surface. Indicates the yaw acceleration and pitch acceleration The total acceleration , Indicates the distance from the acceleration sensor to the vibration suppression surface. represents the multidimensional aerodynamic force measured by the wind tunnel balance, It indicates the distance from the force measuring center of the wind tunnel balance to the vibration suppression surface.

[0063] (4) Based on the principle of torque balance, PID controllers are constructed in the pitch and yaw dimensions respectively. Taking the dynamic load vector constructed in step (3) as the target, the vibration suppression forces to be output in the pitch and yaw dimensions are calculated respectively; then, based on the multi-piezoelectric ceramic actuator collaborative control matrix constructed in step (2) and the amplitude of the dynamic load vector, the input control voltage of each piezoelectric ceramic actuator is solved; finally, each piezoelectric ceramic actuator is driven to output the vibration suppression force to realize the control of the multi-dimensional vibration of the wind tunnel model support system, and a real-time control method for the multi-dimensional vibration of the wind tunnel model support system is established.

[0064] The proposed multi-dimensional vibration control system was built in the laboratory, and the large aspect ratio aircraft model was hammered at an angle of 45°. The pitch and yaw two-dimensional vibration suppression experiments were carried out, and the results were as follows: Figure 7 , Figure 8 As shown, the two-dimensional vibration of the wind tunnel model support system is quickly suppressed.

[0065] This method realizes the effective control of the multi-dimensional vibration of the wind tunnel model support system under complex flow field environment, can characterize the multi-dimensional vibration of the system with high precision and high dynamics, introduces the main control angle to weaken the influence of the stiffness difference between pitch and yaw dimensions on the vibration suppression effect, and realizes the reliable output of multi-dimensional vibration suppression force and the active suppression of multi-dimensional vibration by defining the dynamic load vector and designing a two-dimensional PID controller. It has a good suppression effect on the wind tunnel model support system, can improve the accuracy of aerodynamic measurement data, ensure the reliable implementation of wind tunnel tests, and has the ability to control vibration for various types of wind tunnel models with different multi-dimensional vibration characteristics.

Claims

1. A real-time control method for multi-dimensional vibration of a wind tunnel model support system, characterized in that: Here are the steps: (1) Build a vibration suppression system for the wind tunnel model support system under multi-directional dynamic aerodynamic loads, and simultaneously collect the inertial load and aerodynamic load data of the wind tunnel model support system; (2) In response to the vibration of the wind tunnel model support system caused by multi-directional dynamic aerodynamic loads, a vibration suppressor layout with pitch and yaw dimensions was designed. Four piezoelectric ceramic actuators were evenly distributed along the circumference of the OXY plane to form an "X"-shaped layout. A collaborative control matrix was designed based on the differences in pitch and yaw dimensions of the wind tunnel model. The specific implementation process is as follows: During the vibration suppression process, four piezoelectric ceramic actuators work together to control the vibration caused by multi-directional dynamic aerodynamic loads in the pitch and yaw directions. Due to the differences in dynamic stiffness in pitch and yaw dimensions between different wind tunnel models, the main control angles of the piezoelectric ceramic actuators under different vibration states are defined for different wind tunnel models and their frequency-varying processes. , and define an orthogonal base coordinate system OXY with the center of the actuating section of the piezoelectric ceramic actuator as the origin, and the yaw and pitch dimensions as the X and Y axes, and a non-orthogonal base coordinate system OX1Y1 with the center of the actuating section of the same piezoelectric ceramic actuator as the origin, and the X1 and Y1 axes determined by the main control direction. The conversion relationship between the two is as follows: （1） in, is the coordinate of the external load force orthogonal base coordinate system OXY, is the coordinate in the non-orthogonal base coordinate system OX1Y1; thus, the aerodynamic load of the orthogonal base coordinate system is converted to the non-orthogonal base coordinate system. When the piezoelectric ceramic actuator outputs the response vibration suppression force according to the converted external load, it is equivalent to the pitch and yaw dimensions of the wind tunnel model support system being in a state of equal stiffness. The control quantity of the piezoelectric ceramic actuator with an "X"-shaped inclined 45° layout is directly solved and output without further processing; The coordinated control matrix of multiple piezoelectric ceramic actuators is derived as follows: They represent the control quantity of the piezoelectric ceramic actuator respectively; （2）； (3) filtering the pneumatic load and inertial load data collected in step (1); and transmitting the dynamic load data including the pneumatic load and the inertial load in the target frequency domain to the vibration suppression actuating surface, which is equivalent to the dynamic load vector on the vibration suppression actuating surface; (4) Based on the principle of torque balance, PID controllers are constructed in the pitch and yaw dimensions respectively. Taking the dynamic load vector constructed in step (3) as the target, the vibration suppression forces to be output in the pitch and yaw dimensions are calculated respectively. Then, based on the coordinated control matrix of the multi-piezoelectric ceramic actuator constructed in step (2) and the amplitude of the dynamic load vector, the input control voltage of each piezoelectric ceramic actuator is solved. Finally, each piezoelectric ceramic actuator is driven to output the vibration suppression force to realize the control of the multi-dimensional vibration of the wind tunnel model support system.

2. The multi-dimensional vibration real-time control method of the wind tunnel model support system according to claim 1 is characterized in that: In step (1), a vibration suppression system for a wind tunnel model support system under multi-directional dynamic aerodynamic loads is constructed, which includes an aircraft wind tunnel model, an acceleration sensor, a wind tunnel balance, a tail support rod, a vibration suppressor and a fixing mechanism; two acceleration sensors are respectively arranged in the pitch and yaw dimensions at the center of mass of the aircraft wind tunnel model, and synchronously collect the inertial load and aerodynamic load data of the wind tunnel model support system with the wind tunnel balance, thereby realizing high-precision and high-dynamic collection of multi-directional dynamic aerodynamic loads.

3. The multi-dimensional vibration real-time control method of the wind tunnel model support system according to claim 1 is characterized in that: The specific implementation process of step (3) is as follows: (3.1) Based on the vibration suppression system built in step (1), four piezoelectric ceramic actuators are assembled, the power cord is connected to input the control voltage, and the preload components are matched by wedge surfaces to complete the assembly of the vibration suppression structure; (3.2) Four piezoelectric ceramic actuators are used to output a 0~x Hz sinusoidal sweep signal, which drives the wind tunnel model support system to vibrate, and the time domain response acceleration of the wind tunnel model support system is collected through an acceleration sensor; based on the collected time domain response acceleration data processed by fast Fourier transform, a power spectrum density diagram of the time domain response acceleration is drawn, and the first extreme value of the power spectrum density is found from 0 Hz, and its corresponding frequency is the first-order natural frequency of the wind tunnel model support system; (3.3) A filter is constructed based on the Hilbert-Huang transform to process the collected aerodynamic load and inertial load data and remove the noise signals outside the frequency domain of the first-order natural frequency: First, based on empirical mode decomposition, the upper and lower envelopes of the collected pneumatic load and inertial load data are taken, and the average values ​​of the corresponding points of the upper and lower envelopes are connected into a line, which is the average line; the average line is subtracted from the original load data to obtain a new signal, which is an intrinsic mode function IMF. Repeat this process to decompose the pneumatic load and inertial load signals into several IMFs, and each IMF corresponds to a frequency signal; Then, each IMF is Hilbert transformed as follows, and the instantaneous frequency, amplitude and phase of each order of the mixed signal are obtained after the transformation, and then the time-frequency spectrum of the frequency signal is constructed; （3） in, represents the time domain, represents the real signal before transformation, represents the result of Hilbert transform, Express Perform a Hilbert transform, represents the Cauchy principal components, represents the integral variable, Represents the original signal exist The value of the moment, Indicates the current time point And historical time point The time difference, Represents a and The analytical signal composed of represents the instantaneous amplitude of the signal, represents the instantaneous phase of the signal, represents the instantaneous frequency of the signal, express about The derivative of express about The derivative of Finally, the maximum frequencies corresponding to the instantaneous amplitudes in the frequency signals of each order are obtained after the empirical mode decomposition in the pitch and yaw dimensions respectively as the first-order natural frequencies of the wind tunnel model support system in the pitch and yaw dimensions. At the same time, the signals other than the first-order natural frequencies of the vibration in the pitch and yaw dimensions are eliminated. Finally, a signal filter based on the Hilbert-Huang transform is constructed to eliminate the frequency domain noise interference signals other than the first-order natural frequencies of the wind tunnel model support system. (3.4) The filtered pneumatic load and inertial load data are converted to the vibration suppression surface according to the following formula, and are jointly equivalent to a dynamic load vector; （4） in, It represents the magnitude of the dynamic load vector on the vibration suppression surface. represents the equivalent mass of the cantilever support, wind tunnel balance and wind tunnel model before the vibration suppression surface. Indicates the yaw acceleration and pitch acceleration The total acceleration , Indicates the distance from the acceleration sensor to the vibration suppression surface. represents the multidimensional aerodynamic force measured by the wind tunnel balance, It indicates the distance from the force measuring center of the wind tunnel balance to the vibration suppression surface.

Images