A dynamic calibration method and device for a model force measurement system

Abstract

The present invention relates to the field of wind tunnel testing technology, and in particular to a method and device for dynamic calibration of a model force measurement system, wherein the method comprises: horizontally suspending a model; obtaining multiple sets of hammer input data and output data of each component by tapping to solve the six-component unit impact response function corresponding to the axial force, normal force and pitching moment; rolling the model 90 degrees, obtaining multiple sets of hammer input data and output data of each component by tapping to solve the six-component unit impact response function corresponding to the lateral force and yaw moment; installing a baffle and tapping the baffle to obtain multiple sets of hammer input data and output data of each component to solve the six-component unit impact response function corresponding to the rolling moment; and obtaining a unit impulse response function matrix representing the corresponding relationship between the output data of each component of the model force measurement system and the actual input dynamic load. The present invention can accurately calibrate the model force measurement system used in the zero-stiffness support aerodynamic measurement technology.

Description

Technical Field

[0001] The present invention relates to the technical field of wind tunnel testing, and in particular to a dynamic calibration method and equipment for a model force measurement system and a zero-rigidity support aerodynamic force measurement method. Background Art

[0002] Wind tunnels are ideal ground-based testing equipment for predicting and evaluating high-Mach number aerodynamic performance. When using a wind tunnel for aerodynamic force measurements, the model to be measured is placed within the wind tunnel flow field. The model's force measurement system acquires the corresponding measurement data, and the aerodynamic loads are calculated. To achieve accurate measurements, the model's force measurement system must be dynamically calibrated before testing to establish a corresponding relationship between the measured data and the aerodynamic loads.

[0003] Compared to the common tail support internal balance aerodynamic measurement technology, zero-stiffness support aerodynamic measurement technology can effectively improve the frequency response and data reliability of aerodynamic signal measurements in wind tunnel tests. However, since the model is in a zero-stiffness support state during the test, accurately calibrating the model force measurement system has become a pressing issue. Summary of the Invention

[0004] The object of the present invention is to address at least some of the above-mentioned deficiencies and to provide a model force measurement system dynamic calibration method and equipment suitable for zero-stiffness support aerodynamic force measurement tests, as well as a zero-stiffness support aerodynamic force measurement method.

[0005] In a first aspect, the present invention provides a dynamic calibration method for a model force measurement system, which is used to calibrate a force measurement system of a model having an embedded accelerometer combination and an offline data acquisition device, comprising the following steps:

[0006] Determine the axial, lateral, and normal directions and center of mass of the model;

[0007] The model is suspended horizontally using a suspension device so that the axial direction and normal direction of the model are located in a horizontal plane. The suspension device includes a frame and two vertical lines, with the length of each vertical line being not less than 1 meter. When the model is suspended horizontally, one vertical line is connected from the frame perpendicular to the axial direction of the model to the position of the head of the model that minimizes interference with the flow field, and the other vertical line is connected from the frame perpendicular to the axial direction of the model to the tail end face of the model. The two vertical lines are parallel and spaced apart, and the upper and lower suspension points are both in the plane formed by the axis of the model and the direction of gravity.

[0008] Taking the position where the axis passes through the model head as the striking point P0, a hammer is used to repeatedly strike the model along the axial direction, obtaining multiple sets of hammer input data and corresponding six-component output data of the model force measurement system, and solving the six-component unit impact response function corresponding to the axial force;

[0009] Taking the position where the normal line through the center of mass passes through one side of the model as the striking point PC, a hammer is used to repeatedly strike the model along the normal direction, multiple sets of hammer input data and corresponding six-component output data of the model force measurement system are obtained, and the six-component unit impact response function corresponding to the normal force is solved;

[0010] Select different locations other than the impact point PC in the horizontal plane as the impact points, use a hammer to repeatedly impact the model along the normal direction, and determine the force arm of the impact point relative to the center of mass. Obtain multiple sets of hammer input data, corresponding force arms, and six-component output data of the model force measurement system, and then solve the six-component unit impact response function corresponding to the pitching moment;

[0011] After rolling the model 90° around the axis, the model is horizontally suspended using a suspension device so that the axial and lateral directions of the model are located in the horizontal plane;

[0012] Taking the position where a horizontal perpendicular line passing through the center of mass and perpendicular to the axis passes through one side of the model as the striking point PC', a hammer is used to repeatedly strike the model laterally, obtaining multiple sets of hammer input data and corresponding six-component output data of the model force measurement system, and solving the six-component unit impact response function corresponding to the lateral force;

[0013] Select different locations in the horizontal plane other than the striking point PC' as striking points, use a hammer to repeatedly strike the model laterally, and determine the moment arm of the striking point relative to the center of mass. Obtain multiple sets of hammer input data, corresponding force arms, and six-component output data of the model force measurement system, and then solve the six-component unit impact response function corresponding to the yaw moment;

[0014] Install a baffle at the position where the vertical line through the centroid passes through the upper or lower side of the model;

[0015] At the baffle, a hammer is used to repeatedly strike the baffle in the horizontal direction, and the force arm of the striking point relative to the center of mass is determined. Multiple sets of hammer input data, corresponding force arms, and six-component output data of the model force measurement system are obtained, and then the six-component unit impact response function corresponding to the rolling moment is solved;

[0016] Based on all the solved unit impulse response functions, the unit impulse response function matrix representing the corresponding relationship between the output data of each component of the model force measurement system and the actual input dynamic load is obtained.

[0017] Optionally, solving the six-component unit impact response function corresponding to the axial force includes:

[0018] The unit impact response function values of each component corresponding to the axial force are calculated by using the least squares deconvolution method, with the input data of the axial hammer as input and the output data of each component of the corresponding model force measurement system as output.

[0019] The unit shock response function values of each component obtained by multiple calculations are averaged to obtain the six-component unit shock response function corresponding to the axial force;

[0020] The method of solving the six-component unit impact response function corresponding to the normal force includes:

[0021] The unit impact response function values of each component corresponding to the normal force are calculated by the least squares deconvolution method using the hammer input data of the normal impact as input and the output data of each component of the corresponding model force measurement system as output.

[0022] The unit impact response function values of each component obtained from multiple calculations are averaged to obtain the six-component unit impact response function corresponding to the normal force;

[0023] Solving the six-component unit impulse response function corresponding to the pitching moment includes:

[0024] Based on each set of hammer input data and corresponding stress arm, determine the corresponding pitching moment;

[0025] The unit impact response function values of each component of the normal force and pitching moment are calculated by the least squares deconvolution method using the hammer input data of the normal impact and the corresponding pitching moment as input and the output data of each component of the corresponding model force measurement system as output.

[0026] The unit impact response function values of each component corresponding to the pitching moment obtained by multiple calculations are averaged to obtain the six-component unit impact response function corresponding to the pitching moment;

[0027] Solving the six-component unit impact response function corresponding to the lateral force includes:

[0028] The unit impact response function values of each component of the lateral force are calculated by using the least squares deconvolution method, with the input data of the lateral hammer as input and the output data of each component of the corresponding model force measurement system as output.

[0029] The unit impact response function values of each component obtained from multiple calculations are averaged to obtain the six-component unit impact response function corresponding to the lateral force;

[0030] Solving the six-component unit impulse response function corresponding to the yaw moment includes:

[0031] Based on each set of hammer input data and corresponding stress arm, the corresponding yaw moment is determined;

[0032] The unit impact response function values of the components of the lateral force and yaw moment are calculated by using the least squares deconvolution method, with the input data of the lateral hammer and the corresponding yaw moment as input and the output data of each component of the corresponding model force measurement system as output.

[0033] The unit shock response function values of each component of the yaw moment corresponding to multiple calculations are averaged to obtain the six-component unit shock response function corresponding to the yaw moment;

[0034] The method of solving the six-component unit impulse response function corresponding to the rolling moment includes:

[0035] Based on each set of hammer input data and corresponding stress arm, the corresponding rolling moment is determined;

[0036] The unit impact response function values of the components of the lateral force and rolling moment are calculated by using the least squares deconvolution method, with the input data of the horizontal hammer and the corresponding rolling moment as input and the output data of each component of the corresponding model force measurement system as output.

[0037] The unit impact response function values of each component corresponding to the rolling moment obtained by multiple calculations are averaged to obtain the six-component unit impact response function corresponding to the rolling moment;

[0038] The expression of the unit impulse response function matrix is:

[0039]

[0040] Among them, V A 、V N 、V Z 、V Mz 、V My and V Mx They represent the output data of the axial force, normal force, lateral force, pitching moment, yaw moment and roll moment components of the model force measurement system respectively; FA, FN, FZ, Mz, My and Mx represent the actual input dynamic loads of the six components of axial force, normal force, lateral force, pitching moment, yaw moment and roll moment respectively; “*” represents convolution operation; h AA 、h AN 、h AZ 、h AMz 、h AMy and h AMx are the unit impact response functions of the axial force, normal force, lateral force, pitching moment, yaw moment, and rolling moment components corresponding to the axial force, respectively; h NA 、h NN 、h NZ 、h NMz 、h NMy and hNMx are the unit impact response functions of the axial force, normal force, lateral force, pitching moment, yaw moment, and rolling moment components corresponding to the normal force, respectively; h ZA 、h ZN 、h ZZ 、h ZMz 、h ZMy and h ZMx are the unit impact response functions of the axial force, normal force, lateral force, pitching moment, yaw moment and rolling moment components corresponding to the lateral force respectively; h MzA 、h MzN 、h MzZ 、h MzMz 、h MzMy and h MzMx are the unit impact response functions of the axial force, normal force, lateral force, pitching moment, yaw moment, and rolling moment components corresponding to the pitching moment, respectively; h MyA 、h MyN 、h MyZ 、h MyMz 、h MyMy and h MyMx are the unit impact response functions of the axial force, normal force, lateral force, pitching moment, yaw moment and rolling moment components corresponding to the yaw moment; h MxA 、h MxN 、h MxZ 、h MxMz 、h MxMy and h MxMx They represent the unit impact response functions of the axial force, normal force, lateral force, pitching moment, yaw moment, and rolling moment components corresponding to the rolling moment, respectively.

[0041] Optionally, selecting different positions other than the striking point PC in the horizontal plane as striking points, repeatedly striking the model multiple times along the normal direction with a hammer, and determining the moment arm of the striking point relative to the center of mass includes:

[0042] Determine the intersection line between the current horizontal plane and one side of the model, set at least three tappable points on the intersection line, and remove any tappable point located at the tapping point PC;

[0043] A hammer is used to repeatedly strike the model along the normal direction, and a strikeable point is randomly selected for each strike, and a corresponding force arm is determined according to the position of the selected strikeable point; wherein, at least one strikeable point located in front of the strike point PC is selected, and at least one strikeable point located behind the strike point PC is selected;

[0044] The method of selecting different positions other than the striking point PC' in the horizontal plane as striking points, repeatedly striking the model laterally with a hammer, and determining the force arm of the striking point relative to the center of mass comprises:

[0045] Determine the intersection line between the current horizontal plane and one side of the model, set at least three tappable points on the intersection line, and remove any tappable point located at the tapping point PC';

[0046] A hammer is used to repeatedly strike the model laterally, randomly selecting a strikeable point for each strike, and determining the corresponding force arm based on the position of the selected strikeable point; wherein, a strikeable point located in front of the strike point PC' is selected at least once, and a strikeable point located behind the strike point PC' is selected at least once.

[0047] Optionally, the step of repeatedly striking the model along the axial direction with a hammer, taking the position where the axis passes through the model head as the striking point P0, comprises the following steps:

[0048] Within a preset axial force range, a hammer is used to strike the stationary model along the axial direction of the model at the striking point P0, which is considered as one strike; within the preset axial force range, the swing of the vertical line of the model after being struck does not exceed 1°;

[0049] After the model returns to a stationary state, repeat the previous step until at least 7 taps are completed;

[0050] The method of repeatedly striking the model along the normal direction of the model with a hammer, taking the position where the normal line of the center of mass passes through one side of the model as the striking point PC, comprises the following steps:

[0051] Within a preset normal force range, a hammer is used to strike the stationary model at the striking point PC along the normal direction of the model, which is considered as one strike; within the preset normal force range, the swing of the vertical line of the model after being struck does not exceed 1°;

[0052] After the model returns to a stationary state, repeat the previous step until at least 7 taps are completed;

[0053] The method of repeatedly striking the model laterally with a hammer, using a horizontal vertical line passing through the center of mass and perpendicular to the axis as striking point PC', comprises the following steps:

[0054] Within the preset lateral force range, a hammer is used to strike the stationary model laterally at the striking point PC', which is considered as one strike; within the preset lateral force range, the vertical line of the model does not swing more than 1° after being struck;

[0055] After the model returns to a stationary state, repeat the previous step until at least 7 taps are completed.

[0056] Optionally, repeatedly striking the model multiple times along the normal direction with a hammer, with a randomly selected strikeable point being struck each time, comprises the following steps:

[0057] Randomly select a strikeable point, and within a preset first force range, strike the stationary model along the normal direction of the model with a hammer, which is considered as one strike; within the preset first force range, the pitch angle of the model does not change by more than 3′ after being struck;

[0058] After the model returns to a stationary state, repeat the previous step until at least 7 taps are completed;

[0059] The method of repeatedly striking the model sideways with a hammer, with a randomly selected strikeable point being struck each time, comprises the following steps:

[0060] A strikeable point is randomly selected, and within a preset second force range, a hammer is used to strike the stationary model laterally along the model, which is regarded as one strike; within the preset second force range, the yaw angle of the model does not change by more than 3′ after being struck;

[0061] After the model returns to a stationary state, repeat the previous step until at least 7 taps are completed;

[0062] The baffle is repeatedly struck multiple times in the horizontal direction using a hammer, including the following steps:

[0063] Within a preset third force application range, a hammer is used to strike the stationary model horizontally at the baffle, which is regarded as one strike; within the preset third force application range, after the model is struck, the rolling angle does not change by more than 3′;

[0064] After the model returns to a stationary state, repeat the previous step until at least 7 taps are completed.

[0065] In the second aspect, the present invention also provides a model force measurement system calibration device, which is used to implement the model force measurement system dynamic calibration method as described in any of the above items, including: a suspension device, a force hammer and a host computer, and the host computer is electrically connected to the force hammer and the accelerometer combination embedded in the model.

[0066] Optionally, the model force measurement system calibration equipment further includes a knocking device; the knocking device includes a simple pendulum structure, and the force hammer serves as the pendulum of the simple pendulum, which is used to be released from a preset height and knock the model at the lowest point.

[0067] In a third aspect, the present invention also provides a zero-stiffness support aerodynamic measurement method.

[0068] Calibrate the force measuring system of the model to be measured using any of the above-mentioned dynamic calibration methods for the force measuring system;

[0069] Placing the model in a wind tunnel flow field in the form of a zero-stiffness support, and obtaining measurement results of all the accelerometers within an effective test time;

[0070] Based on the obtained measurement results of all the accelerometers, the accelerometer combination parameters and the force measurement system calibration results, the actual aerodynamic force exerted on the model in the wind tunnel flow field is calculated.

[0071] Optionally, placing the model in a wind tunnel flow field in the form of a zero-stiffness support and obtaining measurement results of all the accelerometers within the effective test time may include:

[0072] placing the model in a preset posture above a uniform flow field area in the wind tunnel;

[0073] Pure release allows the model to enter the uniform region of the wind tunnel flow field in the form of free fall, thereby being in the wind tunnel flow field in the form of zero-stiffness support during the effective test time.

[0074] Optionally, placing the model in a wind tunnel flow field in the form of a zero-stiffness support and obtaining measurement results of all the accelerometers within the effective test time may include:

[0075] The model is placed in a uniform flow field area of the wind tunnel in a preset posture by suspending or lifting;

[0076] The suspension or lifting structure is released so that the model is placed in the wind tunnel flow field in the form of zero-stiffness support within the effective test time.

[0077] The above technical solution of the present invention has the following advantages: the present invention provides a dynamic calibration method and equipment for a model force measurement system. The present invention adopts a suspension method to simulate a zero-stiffness support measurement state, and realizes the dynamic calibration of the model force measurement system through impact response. It can ensure that the calibration state is consistent with the test state as much as possible, with higher accuracy and reliability, and the method is simple and easy to implement.

[0078] The present invention also provides a zero-stiffness support aerodynamic force measurement method, which first uses the above-mentioned model force measurement system dynamic calibration method to calibrate the model force measurement system, and then places the model in the wind tunnel flow field in the form of a zero-stiffness support for measurement, and finally solves the actual aerodynamic force exerted on the model in the wind tunnel flow field. Compared with the existing tail support internal balance aerodynamic force measurement technology, the measurement device has a more compact structure and higher rigidity, which can reduce the interference of difficult-to-handle low-frequency interference on the model aerodynamic force measurement signal, improve data credibility, and thus more accurately measure millisecond-level instantaneous aerodynamic force signals. BRIEF DESCRIPTION OF THE DRAWINGS

[0079] Figure 1 This is a schematic diagram of the steps of a dynamic calibration method for a model force measurement system according to an embodiment of the present invention;

[0080] Figure 2 1. It is a schematic diagram of the distribution of the knocking points and the knockable points on the surface of a model according to an embodiment of the present invention;

[0081] Figure 3 This is another schematic diagram of the distribution of striking points and possible striking points on the surface of a model according to an embodiment of the present invention;

[0082] Figure 4 This is a schematic diagram of the baffle setting.

[0083] In the figure: 1: model; 2: center of mass; 3: baffle. DETAILED DESCRIPTION

[0084] To make the objectives, technical solutions, and advantages of the embodiments of the present invention more clear, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the accompanying drawings in the embodiments of the present invention. Obviously, the described embodiments are part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without making creative efforts shall fall within the scope of protection of the present invention.

[0085] As mentioned above, in order to achieve accurate measurement, the model force measurement system should be dynamically calibrated before the wind tunnel test in order to establish a corresponding relationship between the measured data and the actual aerodynamic load. Compared with the common tail support internal balance aerodynamic measurement technology, the zero-stiffness support aerodynamic measurement technology can effectively improve the wind tunnel test aerodynamic signal measurement frequency response and data credibility. However, since the model is in a zero-stiffness support state during the test, how to accurately calibrate the model force measurement system has become an urgent problem to be solved. In view of this, the present invention provides a method for calibrating the model force measurement system by simulating a zero-stiffness support measurement state through suspension and realizing impact response.

[0086] like Figure 1 As shown, an embodiment of the present invention provides a dynamic calibration method for a model force measurement system, which is used to calibrate a force measurement system of a model having an embedded accelerometer combination and an offline data acquisition device, and includes the following steps:

[0087] Step 100, determining the axial, lateral and normal directions and the center of mass of the model;

[0088] Step 102: suspending the model horizontally using a suspension device so that the axial direction and the normal direction of the model are both in the horizontal plane;

[0089] The suspension device includes a frame and two vertical lines, each of which is no less than 1 meter long. When the model is suspended horizontally, one vertical line is connected from the frame to the axial direction of the model, perpendicular to the model, to the position of the head of the model where the flow field is least disturbed. The other vertical line is connected from the frame to the axial direction of the model, perpendicular to the model, to the tail end face of the model. The two vertical lines are parallel and spaced apart, and the upper and lower suspension points are both in the plane formed by the axis of the model and the direction of gravity.

[0090] Step 104: Using the location where the axis of the model passes through the head of the model as the striking point P0, repeatedly striking the model along the axial direction with a hammer, obtaining multiple sets of hammer input data and corresponding six-component output data of the model force measurement system, and solving the six-component unit impact response function corresponding to the axial force;

[0091] The unit impact response function can be used to express the correspondence between input and output, that is, the correspondence between the actual input dynamic load and the measurement results of the model force measurement system; the six-component output data, that is, the output data corresponding to the six components of axial force, normal force, lateral force, pitching moment, yaw moment and rolling moment; considering that there may be crosstalk between the six components of the model force measurement system, although only the axial force is input when knocking along the axis at point P0, for the measurement results of the model force measurement system, except for the output data of the axial force component, the output data corresponding to the other five components may also not be zero; the six-component unit impact response function corresponding to the axial force, that is, the unit impact response function corresponding to the six components of the axial force, namely, the axial force, normal force, lateral force, pitching moment, yaw moment and rolling moment;

[0092] Step 106: Using the position where the normal line through the center of mass passes through one side of the model as the striking point PC, repeatedly striking the model along the normal direction with a force hammer, obtaining multiple sets of force hammer input data and corresponding six-component output data of the model force measurement system, and solving the six-component unit impact response function corresponding to the normal force;

[0093] The normal through the center of mass is the normal that passes through the center of mass. The impact point PC can be on the left or right side of the model in the current state.

[0094] Step 108: Select different locations in the horizontal plane other than the striking point PC as striking points, repeatedly strike the model multiple times along the normal direction with a force hammer, determine the moment arm of the striking point relative to the center of mass, obtain multiple sets of hammer input data, corresponding force arms, and six-component output data of the model force measurement system, and solve the six-component unit impact response function corresponding to the pitching moment;

[0095] When hitting a position outside the PC point along the normal direction, the model is subjected to both normal force and pitching moment;

[0096] Step 110: After rolling the model 90° around the axis, the model is horizontally suspended using a suspension device so that the axial and lateral directions of the model are located in a horizontal plane.

[0097] Step 112: Using a horizontal perpendicular line passing through the center of mass and perpendicular to the axis and passing through one side of the model as the striking point PC', repeatedly strike the model laterally with a hammer to obtain multiple sets of hammer input data and corresponding six-component output data of the model force measurement system, and solve the six-component unit impact response function corresponding to the lateral force;

[0098] The horizontal vertical line is located in the horizontal plane, perpendicular to the axis of the model, and passes through the center of mass. The striking point PC' can be on the left or right side of the model in the current state.

[0099] Step 114: Select different locations in the horizontal plane other than the striking point PC' as striking points, repeatedly strike the model laterally with a hammer, and determine the moment arms of the striking points relative to the center of mass. Multiple sets of hammer input data, corresponding moment arms, and six-component output data of the model force measurement system are obtained, thereby solving the six-component unit impact response function corresponding to the yaw moment.

[0100] Step 116, installing a baffle at a position where the vertical line through the centroid passes through the upper side or the lower side of the model;

[0101] The vertical line passing through the center of mass is the vertical line passing through the center of mass. The baffle is installed at the top or bottom of the model in the current state.

[0102] Step 118: repeatedly strike the baffle plate multiple times in the horizontal direction with a hammer, determine the moment arm of the striking point relative to the center of mass, obtain multiple sets of hammer input data, corresponding force arms, and six-component output data of the model force measurement system, and then solve the six-component unit impact response function corresponding to the rolling moment;

[0103] Step 120 : Based on all the solved unit impulse response functions, a unit impulse response function matrix is obtained, which represents the corresponding relationship between the output data of each component of the model force measurement system and the actual dynamic load input to the model.

[0104] Among them, the six-component output data of the model force measurement system, namely axial force, normal force, lateral force, pitch moment, yaw moment and roll moment, are determined according to the accelerometer combination and accelerometer combination parameters embedded in the model. The accelerometer combination parameters include the final measurement parameters, installation position, measurement range and measurement accuracy corresponding to each of the accelerometers. The output data of each component of the model force measurement system can be considered as the result obtained by synthesizing and solving the corresponding accelerometer signals.

[0105] An embodiment of the present invention provides a dynamic calibration method for a model force measurement system. The dynamic calibration method uses a suspension method to simulate a zero-stiffness support measurement state. During normal calibration, the normal is placed on a horizontal plane to make the normal approximately in a free motion state. During lateral calibration, the lateral is placed on a horizontal plane to make the lateral approximately in a free motion state. The model is struck by a force hammer, and the force hammer signal is collected as the system input signal and the corresponding accelerometer signal solution result is collected as the system output signal. Then, an intermediate quantity (i.e., a unit impact response function) describing the dynamic response characteristics of the force measurement system is obtained by inverse calculation, thereby realizing calibration of the model force measurement system.

[0106] Optionally, in step 104, solving the six-component unit impact response function corresponding to the axial force further includes:

[0107] The input data of the hammer striking the model along the axial direction is used as input and the output data of each component of the corresponding model force measurement system is used as output. The unit impact response function value of each component corresponding to the axial force is calculated by the least squares deconvolution method.

[0108] The unit shock response function values of each component obtained by multiple calculations are averaged to obtain the six-component unit shock response function corresponding to the axial force;

[0109] In step 106, solving the six-component unit impact response function corresponding to the normal force includes:

[0110] The unit impact response function values of each component corresponding to the normal force are calculated by the least squares deconvolution method using the hammer input data of the normal impact as input and the output data of each component of the corresponding model force measurement system as output.

[0111] The unit impact response function values of each component obtained from multiple calculations are averaged to obtain the six-component unit impact response function corresponding to the normal force;

[0112] In step 108, solving the six-component unit impulse response function corresponding to the pitching moment includes:

[0113] Based on each set of hammer input data and corresponding stress arm, determine the corresponding pitching moment;

[0114] The unit impact response function values of each component of the normal force and pitching moment are calculated by the least squares deconvolution method using the hammer input data of the normal impact and the corresponding pitching moment as input and the output data of each component of the corresponding model force measurement system as output.

[0115] The unit impact response function values of each component corresponding to the pitching moment obtained by multiple calculations are averaged to obtain the six-component unit impact response function corresponding to the pitching moment;

[0116] In step 112, solving the six-component unit impact response function corresponding to the lateral force includes:

[0117] The unit impact response function values of each component of the lateral force are calculated by using the least squares deconvolution method, with the input data of the lateral hammer as input and the output data of each component of the corresponding model force measurement system as output.

[0118] The unit impact response function values of each component obtained from multiple calculations are averaged to obtain the six-component unit impact response function corresponding to the lateral force;

[0119] In step 114, solving the six-component unit impulse response function corresponding to the yaw moment includes:

[0120] Based on each set of hammer input data and corresponding stress arm, the corresponding yaw moment is determined;

[0121] The unit impact response function values of the components of the lateral force and yaw moment are calculated by using the least squares deconvolution method, with the input data of the lateral hammer and the corresponding yaw moment as input and the output data of each component of the corresponding model force measurement system as output.

[0122] The unit shock response function values of each component of the yaw moment corresponding to multiple calculations are averaged to obtain the six-component unit shock response function corresponding to the yaw moment;

[0123] In step 118, solving the six-component unit impulse response function corresponding to the rolling moment includes:

[0124] Based on each set of hammer input data and corresponding stress arm, the corresponding rolling moment is determined;

[0125] The unit impact response function values of the components of the lateral force and rolling moment are calculated by using the least squares deconvolution method, with the input data of the horizontal hammer and the corresponding rolling moment as input and the output data of each component of the corresponding model force measurement system as output.

[0126] The unit impact response function values of each component corresponding to the rolling moment obtained by multiple calculations are averaged to obtain the six-component unit impact response function corresponding to the rolling moment;

[0127] In step 120, the expression of the unit impulse response function matrix is obtained as follows:

[0128]

[0129] Among them, V A 、V N 、V Z 、V Mz 、VMy and V Mx They represent the output data of the axial force, normal force, lateral force, pitching moment, yaw moment and roll moment components of the model force measurement system respectively; FA, FN, FZ, Mz, My and Mx represent the actual input dynamic loads (i.e., the corresponding inputs) of the six components of axial force, normal force, lateral force, pitching moment, yaw moment and roll moment respectively; “*” represents the convolution operation; h AA 、h AN 、h AZ 、h AMz 、h AMy and h AMx They represent the unit shock response functions of the axial force, normal force, lateral force, pitching moment, yaw moment, and rolling moment components corresponding to the axial force (i.e., the six-component unit shock response function corresponding to the axial force); h NA 、h NN 、h NZ 、h NMz 、h NMy and h NMx They represent the unit impact response functions of the axial force, normal force, lateral force, pitching moment, yaw moment, and rolling moment components corresponding to the normal force (i.e., the six-component unit impact response function corresponding to the normal force); h ZA 、h ZN 、h ZZ 、h ZMz 、h ZMy and h ZMx They represent the unit impact response functions of the axial force, normal force, lateral force, pitching moment, yaw moment, and rolling moment components corresponding to the lateral force (i.e., the six-component unit impact response function corresponding to the lateral force); h MzA 、h MzN 、h MzZ 、h MzMz 、h MzMy and h MzMx They represent the unit impulse response functions of the axial force, normal force, lateral force, pitching moment, yaw moment, and rolling moment components corresponding to the pitching moment (i.e., the six-component unit impulse response function corresponding to the pitching moment); h MyA 、h MyN 、h MyZ 、h MyMz 、h MyMy and h MyMx They represent the unit impact response functions of the axial force, normal force, lateral force, pitching moment, yaw moment, and rolling moment components corresponding to the yaw moment (i.e., the six-component unit impact response function corresponding to the yaw moment); h MxA 、h MxN 、h MxZ 、h MxMz 、hMxMy and h MxMx They represent the unit impact response functions of the axial force, normal force, lateral force, pitching moment, yaw moment and rolling moment components corresponding to the rolling moment respectively (i.e., the six-component unit impact response function corresponding to the rolling moment).

[0130] The above matrix is a complete unit impulse response function matrix that takes into account the cross-dimensional interference. When knocking, it is necessary to collect the output of the six components at the same time, and then the knocking component output is deconvolved with the knocking component load input to obtain the unit impulse response function of the main component. For example, when knocking axially, the axial component output data of each component output data of the model force measurement system is deconvolved with the knocking force (that is, the actual input dynamic load) to obtain h AA The output of the other components corresponds to the current knock component load input deconvolution operation to obtain the unit impulse response function of the cross component. For example, the normal component output data when knocking axially is deconvolved with the knock force to obtain h AN The calculation method of other unit impulse response functions is similar.

[0131] The above embodiment uses the least squares deconvolution method to inversely calculate the unit impulse response function value and calculate the average. Calculating the average value is helpful to reduce the random errors generated during the calibration process and obtain a more accurate and reliable calibration result. It should be noted that in step 108, since the position outside the PC point is struck along the normal direction, the model is simultaneously subjected to the normal force and the pitching moment. The input of each calculation includes the hammer input data of the normal strike and the corresponding pitching moment. The unit impact response function values of each component corresponding to the normal force and the pitching moment can be calculated by inversion using the least squares deconvolution method. In theory, the unit impact response function values of each component corresponding to the normal force obtained by multiple calculations are averaged, and the result obtained should be the same as the result in step 106 (i.e., the unit impact response function of each component corresponding to the normal force obtained in step 106). If a large deviation occurs, the dynamic calibration of the model force measurement system should be interrupted and the model force measurement system should be rechecked. Similarly, in steps 114 and 118, the input also includes the lateral force. The unit impact response function values of each component corresponding to the lateral force obtained by inversion calculation are averaged. The result obtained should theoretically be the same as the result in step 112. If a large deviation occurs, the dynamic calibration of the model force measurement system should be interrupted and the model force measurement system should be rechecked.

[0132] Optionally, in step 102, horizontally suspending the model using a suspension device includes:

[0133] Adjusting the axial direction of the frame to be horizontal;

[0134] Adjusting the axial direction and the normal direction of the model to be horizontal, and fixing the model by a temporary support device;

[0135] A vertical line is connected from the frame perpendicular to the axial direction of the model to the head of the model where the flow field is least disturbed, and another vertical line is connected from the frame perpendicular to the axial direction of the model to the tail end face of the model;

[0136] Remove the temporary support device.

[0137] Accordingly, in step 110, the model is horizontally suspended using a suspension device, and the temporary support device may also be used for assistance, including:

[0138] Fixing the model by the temporary support device;

[0139] Remove the hanging thread;

[0140] Roll the model 90° around the axis by using the temporary support device to make the side of the model horizontal;

[0141] A vertical line is connected from the frame perpendicular to the axial direction of the model to the head of the model where the flow field is least disturbed, and another vertical line is connected from the frame perpendicular to the axial direction of the model to the tail end face of the model;

[0142] Remove the temporary support device.

[0143] For models with large volume, weight and inertia, a temporary support device is used to assist in suspending the model, which is conducive to quickly suspending the model horizontally and facilitating the adjustment of the model's posture, ensuring that the two vertical lines are parallel and spaced apart and that the upper and lower hanging points are both in the plane formed by the axis of the model and the direction of gravity.

[0144] Optionally, in step 108, “selecting different positions other than the striking point PC in the horizontal plane as striking points, repeatedly striking the model multiple times along the normal direction with a hammer, and determining the moment arm of the striking point relative to the center of mass” further includes:

[0145] Determine the intersection line between the current horizontal plane and one side of the model, set at least three tappable points on the intersection line, and remove any tappable point located at the tapping point PC;

[0146] A hammer is used to repeatedly strike the model along the normal direction, and a strikeable point is randomly selected for each strike, and a corresponding force arm is determined according to the position of the selected strikeable point; wherein, at least one strikeable point located in front of the strike point PC is selected, and at least one strikeable point located behind the strike point PC is selected;

[0147] The front side is the side relatively close to the head of the model, and the back side is the side relatively close to the tail of the model;

[0148] The step 114 of "selecting different positions other than the striking point PC' in the horizontal plane as striking points, repeatedly striking the model laterally with a hammer, and determining the moment arm of the striking point relative to the center of mass" further includes:

[0149] Determine the intersection line between the current horizontal plane and one side of the model, set at least three tappable points on the intersection line, and remove any tappable point located at the tapping point PC';

[0150] A hammer is used to repeatedly strike the model laterally, randomly selecting a strikeable point for each strike, and determining the corresponding force arm based on the position of the selected strikeable point; wherein, a strikeable point located in front of the strike point PC' is selected at least once, and a strikeable point located behind the strike point PC' is selected at least once.

[0151] The above embodiment randomly selects knocking points in front and behind the knocking point PC for knocking to determine the six-component unit impact response function corresponding to the pitching moment. When knocking at the knocking point PC, the model is only affected by the normal force and no pitching moment is applied. When knocking in the area outside the PC point, the model rotates around the center of mass in the free motion state, and the model is simultaneously affected by the normal force and the pitching moment. Similarly, the six-component unit impact response function corresponding to the yaw moment is determined by randomly selecting knocking points in front and behind the knocking point PC' for knocking. When knocking at the knocking point PC', the model is only affected by the lateral force and no yaw moment is applied. When knocking in the area outside the PC' point, the model rotates around the center of mass in the free motion state, and the model is simultaneously affected by the lateral force and the yaw moment. By knocking randomly multiple times, the accuracy of solving the unit impact response function can be improved.

[0152] Furthermore, at least three tappable points are set on the intersection line, including:

[0153] At least 3 and at most 9 tappable points are evenly arranged on the intersection line.

[0154] The above embodiment uses evenly distributed tappable points on one side of the model to perform repeated random tapping. Tapping different positions is beneficial to measuring the overall sensing effect of the model force measurement system. Too many tappable points will increase the amount of calculation.

[0155] Optionally, in step 104, “taking the position where the axis passes through the head of the model as the striking point P0, repeatedly striking the model along the axial direction with a hammer” includes the following steps:

[0156] Within a preset axial force range, a hammer is used to strike the stationary model along the axial direction of the model at the striking point P0, which is considered as one strike; within the preset axial force range, the swing of the vertical line of the model after being struck does not exceed 1°;

[0157] After the model returns to a stationary state, repeat the previous step until at least 7 taps are completed;

[0158] In step 106, "taking the position where the normal line of the center of mass passes through one side of the model as the striking point PC, repeatedly striking the model multiple times along the normal direction of the model with a hammer" includes the following steps:

[0159] Within a preset normal force range, a hammer is used to strike the stationary model at the striking point PC along the normal direction of the model, which is considered as one strike; within the preset normal force range, the swing of the vertical line of the model after being struck does not exceed 1°;

[0160] After the model returns to a stationary state, repeat the previous step until at least 7 taps are completed;

[0161] In step 112, "taking the position where a horizontal line passing through the center of mass and perpendicular to the axis passes through one side of the model as the striking point PC', repeatedly striking the model sideways with a hammer multiple times" includes the following steps:

[0162] Within the preset lateral force range, a hammer is used to strike the stationary model laterally at the striking point PC', which is considered as one strike; within the preset lateral force range, the vertical line of the model does not swing more than 1° after being struck;

[0163] After the model returns to a stationary state, repeat the previous step until at least 7 taps are completed.

[0164] This method simulates the zero-stiffness support test state of the model in a suspended state. If the model swings significantly, the stress conditions will change, and the corresponding forces and moments will also fluctuate. The measured components can no longer be approximated as a free state. Therefore, the force applied to the model should ensure that the vertical line does not shift significantly. By repeating the tapping at least seven times, obtaining the corresponding input and output data, and ultimately solving for the average value, random errors can be effectively reduced.

[0165] Furthermore, in step 108, "using a hammer to repeatedly strike the model along the normal direction, and randomly selecting a strikeable point for each strike" includes the following steps:

[0166] Randomly select a strikeable point, and within a preset first force range, strike the stationary model along the normal direction of the model with a hammer, which is considered as one strike; within the preset first force range, the pitch angle of the model does not change by more than 3′ after being struck;

[0167] After the model returns to a stationary state, repeat the previous step until at least 7 taps are completed;

[0168] In step 114, "using a hammer to repeatedly strike the model sideways multiple times, and randomly selecting a strikeable point for each strike" includes the following steps:

[0169] A strikeable point is randomly selected, and within a preset second force range, a hammer is used to strike the stationary model laterally along the model, which is regarded as one strike; within the preset second force range, the yaw angle of the model does not change by more than 3′ after being struck;

[0170] After the model returns to a stationary state, repeat the previous step until at least 7 taps are completed;

[0171] In step 118, "repeatedly striking the baffle plate multiple times in a horizontal direction with a hammer" includes the following steps:

[0172] Within a preset third force application range, a hammer is used to strike the stationary model horizontally at the baffle, which is regarded as one strike; within the preset third force application range, after the model is struck, the rolling angle does not change by more than 3′;

[0173] After the model returns to a stationary state, repeat the previous step until at least 7 taps are completed.

[0174] The zero-stiffness support test state of the model is simulated in a suspended state. If the model posture changes significantly, the force situation will also change, and the corresponding forces and moments will also change. The measured components can no longer be approximated as a free state. Therefore, the force applied to the model should ensure that the model posture does not change significantly.

[0175] Taking the pointed cone model as an example, Figure 2 A schematic diagram showing the distribution of knocking points and knockable points on the surface of a model is shown. Figure 2 It is a cross-sectional view, in which direction A represents the axis direction of model 1, and direction N represents the normal direction of model 1. Both directions A and N are located in the horizontal plane. The position where the axis of model 1 passes through the head of model 1 is the striking point P0, and the position where the normal line through the center of mass 2 passes through the side of model 1 is the striking point PC. The striking points P1 to P7 are distributed along one side of model 1, and are respectively set in front of and behind the PC point. The dynamic calibration method of the model force measurement system provided by the present invention is used for calibration. When the axial force component is calibrated in step 104, the striking is repeated multiple times in direction A. Figure 2 At point P0 shown in the figure, record the hammer input data and the output data of each component synthesized by the accelerometer signal, calculate the six-component unit impact response function corresponding to the axial force by the least square deconvolution method and average it. Step 106 When calibrating the normal force component, repeatedly hit the hammer in the N direction for many times. Figure 2At the PC point shown, record the hammer input data and the output data of each component synthesized by the accelerometer signal, calculate the six-component unit impact response function corresponding to the normal force by the least squares deconvolution method and average it. When calibrating the pitch moment component in step 108, repeat random tapping in the N direction. Figure 2 As shown in the figure, at points P1 to P7, record the hammer input data and multiply it by the actual impact point relative to the center of mass 2 to convert it into a torque pulse input signal. Record the output data of each component of the accelerometer signal synthesis. Calculate the six-component unit impact response function corresponding to the pitching moment using the least squares deconvolution method and average it. To calibrate the lateral force component, first rotate the model 1 90° and hang it. At this time, the Z direction and A direction of the lateral direction of the model 1 are both in the horizontal plane. The schematic diagram of the distribution of the impact points and the impactable points on the model surface is shown in the figure. Figure 3 As shown, the position where a horizontal vertical line passing through one side of the model and perpendicular to the axis is taken as the striking point PC', and the striking points P1'~P7' are distributed along one side of the model 1, and are set in front and behind the PC' point. Then repeatedly strike the PC' point multiple times in the Z direction, record the hammer input data and the output data of each component synthesized by the accelerometer signal, calculate the six-component unit impact response function corresponding to the lateral force by the least squares deconvolution method and average it. When calibrating the yaw moment component in step 114, repeatedly and randomly strike the P1'~P7' points in the Z direction, record the hammer input data and multiply it by the length of the force arm of the actual striking point relative to the center of mass 2 to convert it into a torque pulse input signal, record the output data of each component synthesized by the accelerometer signal, calculate the six-component unit impact response function corresponding to the yaw moment by the least squares deconvolution method and average it. As shown Figure 4 As shown, when calibrating the rolling moment component, first install a baffle 3 on the surface of the upper or lower side of the model, then repeatedly strike the baffle 3 in the Z direction, record the hammer input data and multiply it by the arm length of the striking point (i.e., baffle 3) relative to the center of mass 2 to convert it into a torque pulse input signal, record the output data of each component synthesized by the accelerometer signal, calculate the six-component unit impulse response function corresponding to the rolling moment by the least squares deconvolution method and average it. Finally, a complete unit impulse response function matrix that takes into account inter-dimensional cross-interference can be obtained. When using the model 1 force measurement system to measure aerodynamics, that is, when performing load identification, the output of each component of the model 1 force measurement system is obtained, and then the corresponding aerodynamic load of each component can be obtained through the unit impulse response function matrix.

[0176] The present invention also provides a model force measurement system calibration device for implementing the model force measurement system dynamic calibration method described in any of the above embodiments. The calibration device includes: a suspension device, a force hammer, and a host computer, the host computer being electrically connected to the force hammer and the accelerometer embedded in the model. The host computer is configured to collect and record input data from the force hammer and corresponding output data from the model force measurement system, and to calculate a unit impulse response function matrix representing the corresponding relationship between the output data of each component of the model force measurement system and the actual input dynamic load.

[0177] The above embodiment provides a model force measurement system calibration device, which uses a host computer to collect and solve the corresponding data, and uses the aforementioned model force measurement system dynamic calibration method to solve the unit pulse response function matrix representing the correspondence between the output data of each component of the model force measurement system and the actual input dynamic load, which can save manpower and improve computing efficiency.

[0178] Optionally, the calibration device also includes a knocking device; the knocking device includes a simple pendulum structure, and the force hammer serves as the pendulum of the simple pendulum, which is used to be released from a preset height, perform simple pendulum motion along a preset trajectory, and knock the model at the lowest point.

[0179] The motion trajectory of the simple pendulum is fixed. Through the simple pendulum structure, the direction and point of action of the hammer hitting the model can be controlled. Controlling the dynamic striking force mainly involves controlling the time width and amplitude of the impact signal generated by the striking. The amplitude is the peak value of the hammer striking force, and the time width mainly depends on the elastic modulus of the hammer head material. Because ft=mv, the time width t can be controlled by changing the material of the hammer head, and then the size of the striking force peak f can be comprehensively controlled by controlling the striking speed v and the pendulum mass m. By releasing the hammer from different heights, different impact speeds v can be obtained. The preset height (i.e., the starting height of the hammer as the pendulum release) can be calculated based on the force f and speed v required for the striking. By installing weights on the hammer, the pendulum mass m can be changed. The above embodiment can accurately control the size of the striking force by controlling the speed of the hammer striking motion and the hammer's own counterweight, thereby improving the control accuracy of the hammer impact force size and direction, and further realizing accurate calibration of the model force measurement system.

[0180] The present invention also provides a zero-stiffness support aerodynamic measurement method, comprising the following steps:

[0181] Calibrate the force measurement system of the model to be measured using the dynamic calibration method for the model force measurement system as described in any one of the above embodiments;

[0182] Placing the model in a wind tunnel flow field in the form of a zero-stiffness support, and obtaining measurement results of all the accelerometers within an effective test time;

[0183] Based on the measurement results of all the accelerometers, the accelerometer combination parameters and the calibration results of the force measurement system (i.e., the determined unit impulse response function matrix representing the correspondence between the output data of each component of the model force measurement system and the actual input dynamic load), the actual aerodynamic force acting on the model in the wind tunnel flow field is solved.

[0184] The above-described embodiment uses zero-stiffness support to measure aerodynamic forces, making the entire force measurement system more compact and stiffer, thereby significantly increasing the natural vibration frequency and the upper frequency response limit. The method of the present invention has been demonstrated to increase the upper frequency response limit of the force measurement system to above 1 kHz.

[0185] Optionally, placing the model in a wind tunnel flow field in the form of a zero-stiffness support and obtaining measurement results of all the accelerometers within an effective test time further includes:

[0186] placing the model in a preset posture above a uniform flow field area in the wind tunnel;

[0187] Pure release allows the model to enter the uniform region of the wind tunnel flow field in the form of free fall, thereby being in the wind tunnel flow field in the form of zero-stiffness support during the effective test time.

[0188] The above-described embodiment utilizes a pure release method, allowing the model to exist within the uniform region of the wind tunnel flow field in the form of a zero-stiffness support during the effective test time. Pure release refers to a release without initial velocity or acceleration. Because the present invention is suitable for heavy models with large moments of inertia, the model passes through the wind tunnel in a free-fall manner, with minimal changes in attitude, low velocity, and small axial displacement. Compared to the wind tunnel flow field, which can move at several thousand meters per second, the model is equivalent to a stationary state, and the aerodynamic loads it experiences within the uniform region of the wind tunnel flow field can be considered constant. The advantage of pure release is that it has fewer interfering factors on the flow field, but it also places high demands on release control and coordination time with the wind tunnel flow field.

[0189] Optionally, placing the model in a wind tunnel flow field in the form of a zero-stiffness support and obtaining measurement results of all the accelerometers within an effective test time further includes:

[0190] The model is placed in a uniform flow field area of the wind tunnel in a preset posture by suspending or lifting;

[0191] The suspension or lifting structure is released so that the model is placed in the wind tunnel flow field in the form of zero-stiffness support within the effective test time.

[0192] The above-mentioned embodiment adopts the method of suspension release or lifting and withdrawing, so that the model previously placed in the internal area of the wind tunnel test chamber can achieve a zero-rigidity support form. For example, the suspension release can cut the suspension wire or use a fuse or other fusing method to disconnect the suspension wire. After the suspension is released or the lifting is withdrawn, the model falls freely in a predetermined posture inside the wind tunnel. Since the model's displacement in the direction of gravity and axial direction is very small in the initial stage of falling, the model can be approximately hovered in the flow field, and the aerodynamic load on the model in the uniform area of the wind tunnel flow field can be considered to be constant. The advantage of releasing the suspension or lifting is that it is easy to make the model in the ideal position in the uniform area of the wind tunnel flow field within the effective test time window, but the control requirements for the suspension release or lifting and withdrawal are relatively high.

[0193] In summary, the present invention provides a dynamic calibration method and equipment for a model force measurement system. The present invention realizes the simulation of the free state of each component through suspension direction control, making the calibration state closer to the test state. While simulating the free state of each component of the model, the dynamic calibration of each component is realized, which has high accuracy and reliability, is easy to implement, and has low cost.

[0194] The present invention also provides a zero-stiffness support aerodynamic force measurement method, which first uses the above-mentioned model force measurement system dynamic calibration method to calibrate the model force measurement system, and then places the model in the wind tunnel flow field in the form of a zero-stiffness support for measurement, and finally solves the actual aerodynamic force exerted on the model in the wind tunnel flow field. Compared with the existing tail support internal balance aerodynamic force measurement technology, the measurement device has a more compact structure and higher rigidity, which can reduce the interference of difficult-to-handle low-frequency interference on the model aerodynamic force measurement signal, improve data credibility, and thus more accurately measure millisecond-level instantaneous aerodynamic force signals.

[0195] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, rather than to limit it. Although the present invention has been described in detail with reference to the aforementioned embodiments, those skilled in the art should understand that they can still modify the technical solutions described in the aforementioned embodiments, or make equivalent replacements for some of the technical features therein. However, these modifications or replacements do not deviate the essence of the corresponding technical solutions from the spirit and scope of the technical solutions of the various embodiments of the present invention.

Claims

1. A dynamic calibration method for a model force measurement system, characterized in that: The method is used to calibrate the force measurement system of a model having an embedded accelerometer combination and an offline data acquisition device, including the following steps: Determine the axial, lateral, and normal directions and center of mass of the model; The model is suspended horizontally using a suspension device so that the axial direction and normal direction of the model are located in a horizontal plane. The suspension device includes a frame and two vertical lines, with the length of each vertical line being not less than 1 meter. When the model is suspended horizontally, one vertical line is connected from the frame perpendicular to the axial direction of the model to the position of the head of the model that minimizes interference with the flow field, and the other vertical line is connected from the frame perpendicular to the axial direction of the model to the tail end face of the model. The two vertical lines are parallel and spaced apart, and the upper and lower suspension points are both in the plane formed by the axis of the model and the direction of gravity. Taking the position where the axis passes through the model head as the striking point P0, a hammer is used to repeatedly strike the model along the axial direction, obtaining multiple sets of hammer input data and corresponding six-component output data of the model force measurement system, and solving the six-component unit impact response function corresponding to the axial force; Taking the position where the normal line through the center of mass passes through one side of the model as the striking point PC, a hammer is used to repeatedly strike the model along the normal direction, multiple sets of hammer input data and corresponding six-component output data of the model force measurement system are obtained, and the six-component unit impact response function corresponding to the normal force is solved; Select different locations other than the impact point PC in the horizontal plane as the impact points, use a hammer to repeatedly impact the model along the normal direction, and determine the force arm of the impact point relative to the center of mass. Obtain multiple sets of hammer input data, corresponding force arms, and six-component output data of the model force measurement system, and then solve the six-component unit impact response function corresponding to the pitching moment; After rolling the model 90° around the axis, the model is horizontally suspended using a suspension device so that the axial and lateral directions of the model are located in the horizontal plane; Taking the position where a horizontal perpendicular line passing through the center of mass and perpendicular to the axis passes through one side of the model as the striking point PC', a hammer is used to repeatedly strike the model laterally, obtaining multiple sets of hammer input data and corresponding six-component output data of the model force measurement system, and solving the six-component unit impact response function corresponding to the lateral force; Select different locations in the horizontal plane other than the striking point PC' as striking points, use a hammer to repeatedly strike the model laterally, and determine the moment arm of the striking point relative to the center of mass. Obtain multiple sets of hammer input data, corresponding force arms, and six-component output data of the model force measurement system, and then solve the six-component unit impact response function corresponding to the yaw moment; Install a baffle at the position where the vertical line through the centroid passes through the upper or lower side of the model; At the baffle, a hammer is used to repeatedly strike the baffle in the horizontal direction, and the force arm of the striking point relative to the center of mass is determined. Multiple sets of hammer input data, corresponding force arms, and six-component output data of the model force measurement system are obtained, and then the six-component unit impact response function corresponding to the rolling moment is solved; Based on all the solved unit impulse response functions, the unit impulse response function matrix representing the corresponding relationship between the output data of each component of the model force measurement system and the actual input dynamic load is obtained.

2. The dynamic calibration method of a model force measurement system according to claim 1, characterized in that: The six-component unit impact response function corresponding to the axial force is solved, including: The unit impact response function values of each component corresponding to the axial force are calculated by using the least squares deconvolution method, with the input data of the axial hammer as input and the output data of each component of the corresponding model force measurement system as output. The unit shock response function values of each component obtained by multiple calculations are averaged to obtain the six-component unit shock response function corresponding to the axial force; The method of solving the six-component unit impact response function corresponding to the normal force includes: The unit impact response function values of each component corresponding to the normal force are calculated by the least squares deconvolution method using the hammer input data of the normal impact as input and the output data of each component of the corresponding model force measurement system as output. The unit impact response function values of each component obtained from multiple calculations are averaged to obtain the six-component unit impact response function corresponding to the normal force; Solving the six-component unit impulse response function corresponding to the pitching moment includes: Based on each set of hammer input data and corresponding stress arm, determine the corresponding pitching moment; The unit impact response function values of each component of the normal force and pitching moment are calculated by the least squares deconvolution method using the hammer input data of the normal impact and the corresponding pitching moment as input and the output data of each component of the corresponding model force measurement system as output. The unit impact response function values of each component corresponding to the pitching moment obtained by multiple calculations are averaged to obtain the six-component unit impact response function corresponding to the pitching moment; Solving the six-component unit impact response function corresponding to the lateral force includes: The unit impact response function values of each component of the lateral force are calculated by using the least squares deconvolution method, with the input data of the lateral hammer as input and the output data of each component of the corresponding model force measurement system as output. The unit impact response function values of each component obtained from multiple calculations are averaged to obtain the six-component unit impact response function corresponding to the lateral force; Solving the six-component unit impulse response function corresponding to the yaw moment includes: Based on each set of hammer input data and corresponding stress arm, the corresponding yaw moment is determined; The unit impact response function values of the components of the lateral force and yaw moment are calculated by using the least squares deconvolution method, with the input data of the lateral hammer and the corresponding yaw moment as input and the output data of each component of the corresponding model force measurement system as output. The unit shock response function values of each component of the yaw moment corresponding to multiple calculations are averaged to obtain the six-component unit shock response function corresponding to the yaw moment; The method of solving the six-component unit impulse response function corresponding to the rolling moment includes: Based on each set of hammer input data and corresponding stress arm, the corresponding rolling moment is determined; The unit impact response function values of the components of the lateral force and rolling moment are calculated by using the least squares deconvolution method, with the input data of the horizontal hammer and the corresponding rolling moment as input and the output data of each component of the corresponding model force measurement system as output. The unit impact response function values of each component corresponding to the rolling moment obtained by multiple calculations are averaged to obtain the six-component unit impact response function corresponding to the rolling moment; The expression of the unit impulse response function matrix is: Among them, V A 、V N 、V Z 、V Mz 、V My and V Mx They represent the output data of the axial force, normal force, lateral force, pitching moment, yaw moment and roll moment components of the model force measurement system respectively; FA, FN, FZ, Mz, My and Mx represent the actual input dynamic loads of the six components of axial force, normal force, lateral force, pitching moment, yaw moment and roll moment respectively; "*" represents convolution operation; h AA 、h AN 、h AZ 、h AMz 、h AMy and h AMx are the unit impact response functions of the axial force, normal force, lateral force, pitching moment, yaw moment and rolling moment components corresponding to the axial force respectively; h NA 、h NN 、h NZ 、h NMz 、h NMy and h NMx are the unit impact response functions of the axial force, normal force, lateral force, pitching moment, yaw moment, and rolling moment components corresponding to the normal force, respectively; h ZA 、h ZN 、h ZZ 、h ZMz 、h ZMy and h ZMx are the unit impact response functions of the axial force, normal force, lateral force, pitching moment, yaw moment and rolling moment components corresponding to the lateral force respectively; h MzA 、h MzN 、h MzZ 、h MzMz 、h MzMy and h MzMx are the unit impact response functions of the axial force, normal force, lateral force, pitching moment, yaw moment, and rolling moment components corresponding to the pitching moment, respectively; h MyA 、h MyN 、h MyZ 、h MyMz 、h MyMy and h MyMx are the unit impact response functions of the axial force, normal force, lateral force, pitching moment, yaw moment and rolling moment components corresponding to the yaw moment; h MxA 、h MxN 、h MxZ 、h MxMz 、h MxMy and h MxMx They represent the unit impact response functions of the axial force, normal force, lateral force, pitching moment, yaw moment, and rolling moment components corresponding to the rolling moment, respectively.

3. The dynamic calibration method of a model force measurement system according to claim 1, characterized in that: The method of selecting different positions other than the striking point PC in the horizontal plane as striking points, repeatedly striking the model multiple times along the normal direction with a hammer, and determining the force arm of the striking point relative to the center of mass includes: Determine the intersection line between the current horizontal plane and one side of the model, set at least three tappable points on the intersection line, and remove any tappable point located at the tapping point PC; A hammer is used to repeatedly strike the model along the normal direction, and a strikeable point is randomly selected for each strike, and a corresponding force arm is determined according to the position of the selected strikeable point; wherein, at least one strikeable point located in front of the strike point PC is selected, and at least one strikeable point located behind the strike point PC is selected; The method of selecting different positions other than the striking point PC' in the horizontal plane as striking points, repeatedly striking the model laterally with a hammer, and determining the force arm of the striking point relative to the center of mass comprises: Determine the intersection line between the current horizontal plane and one side of the model, set at least three tappable points on the intersection line, and remove any tappable point located at the tapping point PC'; A hammer is used to repeatedly strike the model laterally, randomly selecting a strikeable point for each strike, and determining the corresponding force arm based on the position of the selected strikeable point; wherein, a strikeable point located in front of the strike point PC' is selected at least once, and a strikeable point located behind the strike point PC' is selected at least once.

4. The dynamic calibration method of a model force measurement system according to claim 1, characterized in that: The position where the axis passes through the head of the model is used as the striking point P0, and a hammer is used to repeatedly strike the model along the axial direction, including the following steps: Within a preset axial force range, a hammer is used to strike the stationary model along the axial direction of the model at the striking point P0, which is considered as one strike; within the preset axial force range, the swing of the vertical line of the model after being struck does not exceed 1°; After the model returns to a stationary state, repeat the previous step until at least 7 taps are completed; The method of repeatedly striking the model along the normal direction of the model with a hammer, taking the position where the normal line of the center of mass passes through one side of the model as the striking point PC, comprises the following steps: Within a preset normal force range, a hammer is used to strike the stationary model at the striking point PC along the normal direction of the model, which is considered as one strike; within the preset normal force range, the swing of the vertical line of the model after being struck does not exceed 1°; After the model returns to a stationary state, repeat the previous step until at least 7 taps are completed; The method of repeatedly striking the model laterally with a hammer, using a horizontal vertical line passing through the center of mass and perpendicular to the axis as striking point PC', comprises the following steps: Within the preset lateral force range, a hammer is used to strike the stationary model laterally at the striking point PC', which is considered as one strike; within the preset lateral force range, the vertical line of the model does not swing more than 1° after being struck; After the model returns to a stationary state, repeat the previous step until at least 7 taps are completed.

5. The dynamic calibration method of a model force measurement system according to claim 3, characterized in that: The method of repeatedly striking the model along the normal direction with a hammer, with a randomly selected strikeable point being struck each time, comprises the following steps: Randomly select a strikeable point, and within a preset first force range, use a hammer to strike the stationary model along the normal direction of the model, which is considered as one strike; within the preset first force range, the pitch angle of the model does not change by more than 3′ after being struck; After the model returns to a stationary state, repeat the previous step until at least 7 taps are completed; The method of repeatedly striking the model sideways with a hammer, with a randomly selected strikeable point being struck each time, comprises the following steps: A strikeable point is randomly selected, and within a preset second force range, a hammer is used to strike the stationary model laterally along the model, which is regarded as one strike; within the preset second force range, the yaw angle of the model does not change by more than 3′ after being struck; After the model returns to a stationary state, repeat the previous step until at least 7 taps are completed; The baffle is repeatedly struck multiple times in the horizontal direction using a hammer, including the following steps: Within a preset third force application range, a hammer is used to strike the stationary model horizontally at the baffle, which is regarded as one strike; within the preset third force application range, after the model is struck, the rolling angle does not change by more than 3′; After the model returns to a stationary state, repeat the previous step until at least 7 taps are completed.

6. A model force measurement system calibration device, characterized in that: The method for dynamic calibration of a model force measurement system according to any one of claims 1 to 5 comprises: a suspension device, a force hammer and a host computer, wherein the host computer is electrically connected to the force hammer and the accelerometer embedded in the model.

7. The model force measurement system calibration device according to claim 6, characterized in that: It also includes a striking device; the striking device includes a simple pendulum structure, and the force hammer serves as the pendulum of the simple pendulum, which is used to be released from a preset height and strike the model at the lowest point.

8. A zero-stiffness support aerodynamic measurement method, characterized in that: The steps include: Calibrate the force measuring system of the model to be measured using the dynamic calibration method for the model force measuring system according to any one of claims 1 to 5; Placing the model in a wind tunnel flow field in the form of a zero-stiffness support, and obtaining measurement results of all the accelerometers within an effective test time; Based on the obtained measurement results of all the accelerometers, the accelerometer combination parameters and the force measurement system calibration results, the actual aerodynamic force exerted on the model in the wind tunnel flow field is calculated.

9. The zero-stiffness support aerodynamic force measurement method according to claim 8, characterized in that: The model is placed in the wind tunnel flow field in the form of a zero-stiffness support, and the measurement results of all the accelerometers within the effective test time are obtained, including: placing the model in a preset posture above a uniform flow field area in the wind tunnel; Pure release allows the model to enter the uniform region of the wind tunnel flow field in the form of free fall, thereby being in the wind tunnel flow field in the form of zero-stiffness support during the effective test time.

10. The zero-stiffness support aerodynamic force measurement method according to claim 8, characterized in that: The model is placed in the wind tunnel flow field in the form of a zero-stiffness support, and the measurement results of all the accelerometers within the effective test time are obtained, including: The model is placed in a uniform flow field area of the wind tunnel in a preset posture by suspending or lifting; The suspension or lifting structure is released so that the model is placed in the wind tunnel flow field in the form of zero-stiffness support within the effective test time.

Images