A wind tunnel standard force model longitudinal load in-situ calibration device and method

Abstract

A method and apparatus for in-situ calibration of longitudinal load on a standard wind tunnel force measurement model, belonging to the field of aerodynamic wind tunnel testing technology, addresses the lack of in-situ loading correction methods in wind tunnel testing. The method ensures accurate loading of the standard wind tunnel force measurement model on-site through weight loading and high-precision angle sensor monitoring. By loading the model in-situ, the center distance, fixed installation angle, and elastic installation angle are obtained, thus ensuring accurate conversion of the balance force to the model's axis system and enabling real-world verification of the balance's accuracy under the same axis system. Based on the loading verification results, single-component in-situ calibration is performed on the non-compliant components. The principal coefficients obtained from the in-situ calibration are used to replace the principal coefficients of the non-compliant components, and constantization is applied to establish a new calibration formula model. The balance formula is recalculated using offline calibration data, and iterative corrections are performed until the balance measurement error meets the requirements.

Description

Technical Field

[0001] This invention belongs to the field of aerodynamic wind tunnel testing technology, specifically relating to an in-situ calibration device and method for a wind tunnel standard force measurement model, and more particularly to an in-situ calibration device and method for directly applying a standard force to a wind tunnel standard force measurement model at the wind tunnel test site. Background Technology

[0002] Wind tunnel standard model force measurement tests are an important part of wind tunnel flow field calibration and a crucial basis for evaluating the relevance of wind tunnel tests. Wind tunnel standard force measurement tests generally only perform longitudinal tests to obtain the lift coefficient, drag coefficient, and pitching moment coefficient under specified Mach numbers and angles of attack. Therefore, accurate measurement of lift, drag, and pitching moment is of great significance for wind tunnel standard model force measurement tests.

[0003] The wind tunnel balance (or simply balance) is a key device for measuring the aforementioned aerodynamic loads. The balance formula is obtained through offline static calibration on the ground. However, since the calibration state cannot completely replicate the wind tunnel test state, a "state difference" exists between the calibration data and the test data. This difference can be caused by variations in excitation power supply due to cable issues, differences in the amplification factor of the acquisition system, differences between calibration and test temperatures, and differences in the balance's installation state / force transmission path. These differences result in the balance formula obtained through offline calibration becoming inapplicable when used in the wind tunnel, manifesting as a change in the balance's sensitivity coefficient. If this cannot be effectively corrected, it will directly affect the reliability of the wind tunnel test data.

[0004] However, the current wind tunnel testing field lacks in-situ calibration methods and standards based on wind tunnel test sites, and effective in-situ loading correction methods are scarce. The on-site loading verification methods vary significantly across different wind tunnels and different wind tunnel test types, with a widespread phenomenon of "verification without correction." For wind tunnel standard force model tests that highly focus on longitudinal aerodynamic characteristics, the "state difference" between calibration data and test data is particularly prominent. In such tests, discrepancies frequently arise between test results from different balances in the same wind tunnel. Simply attributing poor correlation in wind tunnel tests to balance performance is insufficient. There is an urgent need to establish in-situ calibration methods and techniques for longitudinal loads on wind tunnel standard force models to improve the reliability of wind tunnel test data. Summary of the Invention

[0005] The purpose of this invention is to provide an in-situ calibration device and method for longitudinal loads on a standard wind tunnel force measurement model, thereby addressing the problem of the lack of in-situ loading correction methods in wind tunnel testing. The technical solution adopted by this invention is as follows:

[0006] A wind tunnel standard force measurement model longitudinal load in-situ calibration device includes a standard model, a balance, a support rod, an angle sensor, and a loading rod. The standard model is connected to the support rod via the balance. A model loading plate is provided on the lower side of the standard model, and an angle sensor is provided on the model loading plate. The upper end of the loading rod is hinged to the model loading plate via a universal joint, and several weights are provided at the bottom of the loading rod.

[0007] Furthermore, the model loading plate is set horizontally, and two vertically intersecting T-slots are provided on the lower end surface of the model loading plate. The universal joint includes a ball seat and a loading ball that are rotatably engaged. The upper end of the loading rod is connected to the loading ball, and the ball seat is slidably set in either T-slot.

[0008] The T-slot has a first scale on one side, with a minimum scale value of 1mm. The ball seat has a second scale, with a minimum scale value of 0.9mm.

[0009] Furthermore, the loading rod is threadedly connected to the loading ball, the standard model is provided with a conical hole, the front end of the balance is provided with a conical head, the conical head is adapted to the conical hole for insertion and engagement, and is tightened and fixed by the first screw, and the model loading plate is connected to the standard model by the second screw.

[0010] This invention also provides an in-situ calibration method for longitudinal loads of a wind tunnel standard force measurement model, which is based on the aforementioned in-situ calibration device for longitudinal loads of a wind tunnel standard force measurement model, and includes the following steps:

[0011] Step 1: Establish the balance formula;

[0012] Step 2: Measure the distance between the two centers;

[0013] Step 3: Measure the installation angle;

[0014] Step 4: Convert the components of the balance to the axis system of the standard model to obtain the converted value of the longitudinal component;

[0015] Step 5: Apply the test load to the standard model to obtain the nominal value of the test load;

[0016] Step 6: Calculate the nominal deviation of the longitudinal component between the converted value and the nominal value, and determine whether the nominal deviation is acceptable. If the nominal deviation is acceptable, the calibration is complete. If the nominal deviation exceeds the tolerance, the principal coefficient is corrected, and the process returns to Step 1 to re-establish the balance formula and iterate the calibration.

[0017] Furthermore, the specific steps of step two are as follows:

[0018] First, remove the loading rod and several weights. Use the angle sensor to level the standard model. Select a loading position at a specified distance from the torque reference point of the standard model. Then, install the loading rod and several weights. Load the standard model at the loading position using the loading rod and several weights. Level the standard model again and read the measured values ​​of the six components of the balance.

[0019] Calculate the distance between the center of the balance and the torque reference point using the following formula:

[0020] ;

[0021] In the formula, Mz b The pitching moment measured for the balance;

[0022] Y b The normal force measured by the balance;

[0023] LP x This is the distance between the torque reference point and the axis of the loading rod.

[0024] Furthermore, the specific steps of step three are as follows:

[0025] Step 31, establish the installation angle formula:

[0026] α = α0 + α t ;

[0027] In the formula, α is the installation angle between the standard model and the balance; α0 is the fixed installation angle between the standard model and the balance when there is no load.

[0028] α t α is the change in the installation angle between the standard model and the balance under load, i.e., the elastic installation angle. t The linear relationship between the load and the load is as follows:

[0029] α t =k Mz Mz;

[0030] In the formula, k Mz This is a correction factor for the flexible installation angle;

[0031] Define the center of the large-diameter end of the cone of the balance as the connection point between the balance and the standard model, and Mz as the torque that transforms the forces and moments in the balance coordinate system to the connection point. Mz is calculated by the following formula:

[0032] Mz=Y b L+Mz b ;

[0033] In the formula, L is the lever arm of the normal force, which is calculated using the following formula:

[0034] L=L x +L0;

[0035] In the formula, L0 is the distance between the torque reference point and the connection point in the x-axis direction in the standard model coordinate system;

[0036] Step 32, loading in both directions:

[0037] With the standard model in its upright position, adjust it to a horizontal position using an angle sensor. Suspend a loading rod and several weights at the loading position, and record the axial force X measured by the balance. b+ and normal force Y b+ And calculate the installation angle α between the standard model and the balance in the upright state according to the following formula. + :

[0038] ;

[0039] With the standard model in an inverted position (back down), attach the loading plate to the back of the standard model. Adjust the standard model to be horizontal using an angle sensor. Suspend the loading rod and several weights at the loading position and record the axial force X measured by the balance. b- and reverse normal force Y b- The installation angle α between the standard model and the balance in the reversed state is calculated according to the following formula. - :

[0040] ;

[0041] Step 33, separate the fixed mounting angle from the flexible mounting angle:

[0042] Taking advantage of the fact that the fixed mounting angle remains unchanged in both the upright and reverse mounting states, while the flexible mounting angle is related to the polarity of Mz, the fixed mounting angle α0 and the flexible mounting angle α are determined. t :

[0043] ;

[0044] ;

[0045] Step 34, calculate the flexible installation angle correction coefficient k Mz :

[0046] .

[0047] Furthermore, the specific steps of step four are as follows:

[0048] Transform the component vectors of the balance scale to the coordinate system of the standard model:

[0049] Y m =Yb cos(α0+k Mz Mz);

[0050] Mz m =Mz b +Y m L x ;

[0051] X m =Y b sin(α0+k Mz Mz);

[0052] In the formula, Y m This is the value of the normal force of the balance converted to the axis system of the standard model;

[0053] Mz m This is the pitch moment of the balance converted to the axis system value of the standard model;

[0054] X m This is the value converted from the axial force of the balance to the shaft system of the standard model.

[0055] Furthermore, the specific steps of step five are as follows:

[0056] The test load is applied to the standard model using a model loading plate. The test load under the axis system of the standard model is expressed as:

[0057] Y N =Gcosα N ;

[0058] Mz N =Y N LP x ;

[0059] X N =Gsinα N ;

[0060] In the formula, Y N The nominal value of the normal force for the test load under the standard model shaft system;

[0061] Mz N The nominal value of the pitching moment under the test load in the standard model shaft system;

[0062] X N The nominal value of the axial force for the test load under the standard model shaft system;

[0063] α N The nominal angle of attack for the standard model is measured by an angle sensor;

[0064] G is the nominal weight of the loading rod and several weights.

[0065] Furthermore, the specific steps of step six are as follows:

[0066] The nominal deviation between the converted value and the nominal value of the corresponding component is:

[0067] ;

[0068] ;

[0069] ;

[0070] In the formula, δ X δ represents the nominal deviation between the converted value and the nominal value of the axial force component. Y δ represents the nominal deviation between the converted value and the nominal value of the normal force component. Mz This represents the nominal deviation between the upconverted value and the nominal value of the pitch moment component.

[0071] Determine δ X δ Y and δ Mz Does it exceed 120% of the balance's accuracy range? If δ X δ Y and δ Mz If the result is within 120% accuracy of the balance, the calibration is complete; otherwise, correct the main coefficient and return to step one to re-establish the balance formula and iterate the calibration.

[0072] Furthermore, the specific steps for correcting the principal term coefficient are as follows:

[0073] Step 61, apply the correction load to the standard model:

[0074] Based on the required distance between the two centers and the fixed installation angle, adjust the standard model to make the axis of the balance horizontal, adjust the axial position of the ball seat, and when the absolute value of the axial force output of the balance is less than or equal to 0.01N, use the current position of the ball seat as the normal force loading correction position, and apply the normal force correction load to the standard model.

[0075] Choose a front and rear span of L. c Two known loading points are used as pitch moment loading correction positions. First, weights of equal weight are hung at the two pitch moment loading correction positions respectively. Then, the weights at the two pitch moment loading correction positions are swapped to reload the pitch moment correction load.

[0076] Stand the balance upright and apply an axial force correction load;

[0077] Step 62, calculate the principal coefficient:

[0078] Calculate the principal coefficient using the following formula:

[0079] ;

[0080] In the formula, The principal coefficient of the i-th component;

[0081] This is the j-th calibration load applied to the i-th component;

[0082] To load the i-th component signal corresponding to the j-th calibration load;

[0083] Step 63, constantization of the principal term coefficients:

[0084] Offline calibration formula Undetermined coefficients in Replace with The general formula for balance calibration becomes:

[0085] ;

[0086] In the formula, F i The load is the i-th component of the balance.

[0087] F j The load of the j-th component that interferes with the i-th component;

[0088] F k The load of the k-th component that interferes with the i-th component;

[0089] The i-th component signal is output.

[0090] The principal coefficient of the i-th component;

[0091] The first-order interference correction coefficient of the j-th component to the i-th component;

[0092] Here are the squared interference correction factors (j=k) or cross-interference correction factors (j≠k) for each component load on the i-th component, with the following units:

[0093] The force-force interference coefficient for force components is expressed in N / N. 2 ;

[0094] The force-force disturbance coefficient for the torque component is expressed in Nm / N. 2 ;

[0095] The force-torque disturbance coefficient for force components is expressed in N / N•Nm.

[0096] The interference coefficient of the force-torque component is expressed in Nm / N•Nm.

[0097] The torque-torque disturbance coefficient of the force component is expressed in N / Nm. 2 ;

[0098] The cross-interference coefficient of torque-torque components is expressed in Nm / Nm. 2 ;

[0099] The force-force cross-term interference coefficient for the torque component is expressed in Nm / N. 2 ;

[0100] At this point, the coefficient of the principal term becomes a constant;

[0101] Step 64: Recalculate the balance formula using the offline calibration data:

[0102] Calling the voltage signal output matrix during ground offline calibration [ U] and calibration load matrix [P]:

[0103] ;

[0104] ;

[0105] In the formula, m is the total number of loading groups. For a six-component balance, m = 144.

[0106] The least squares model for calculating the balance formula is as follows:

[0107] ;

[0108] In the formula, q is the loading group number, with a total of m groups, where m is greater than the number of undetermined coefficients. The voltage signal output for the i-th element when the q-th group is loaded; The load of the i-th element when the q-th group is loaded; The load of the j-th element when the q-th group is loaded; The load of the k-th element when the q-th group is loaded;

[0109] Solve the following system of equations to obtain the undetermined coefficients. , :

[0110] ;

[0111] ;

[0112] After obtaining the new formula for the balance scale, return to step one. Once the calculated load accuracy reaches the nominal accuracy of the balance, the flat-state loading correction of the balance is complete; otherwise, continue iterative calculation.

[0113] Compared with the prior art, the beneficial effects of the present invention are as follows:

[0114] 1. This invention proposes to perform in-situ loading on a standard model in a wind tunnel and correct the loading error by calibrating the main term coefficient, which solves the problem that measurement errors detected in current wind tunnel tests cannot be corrected in-situ.

[0115] 2. This invention considers the influence of the connection stiffness between the balance and the standard model on the measurement, introduces the elastic installation angle correction coefficient of the balance and the standard model, establishes the installation angle and load influence model and the elastic installation angle calculation method, solves the problem of lack of correction means for the installation angle of the balance and the standard model, and improves the accuracy of the force and moment coordinate conversion of the balance.

[0116] 3. This invention proposes a method for recalculating the balance formula coefficients by constantizing the main term coefficients. By calling back the original offline calibration data, the balance formula coefficients are recalculated, thereby improving the accuracy of wind tunnel balance load prediction. Attached Figure Description

[0117] Figure 1 This is a schematic diagram of the calibration device of the present invention;

[0118] Figure 2 This is a bottom view of the model loading plate;

[0119] Figure 3 It is a cross-sectional view of the universal joint engaging with the model loading plate;

[0120] Figure 4 This is the loading diagram of the standard model in its normal loading state;

[0121] Figure 5 This is the loading diagram of the inverted state of the standard model;

[0122] Figure 6 This is a flowchart of the calibration method of the present invention.

[0123] In the diagram, 1. Standard model, 2. First screw, 3. Balance, 4. Support rod, 5. Model loading plate, 6. Angle sensor, 7. Second screw, 8. Loading rod, 9. Weight, 10. Universal joint, 11. Ball seat, 12. Loading ball, 13. First scale, 14. Model center of gravity, 15. Balance center, 16. T-slot, 17. Second scale. Detailed Implementation

[0124] To make the objectives, technical solutions, and advantages of this invention clearer, the invention is described below with reference to specific embodiments shown in the accompanying drawings. However, it should be understood that these descriptions are merely exemplary and not intended to limit the scope of the invention. Furthermore, descriptions of well-known structures and technologies are omitted in the following description to avoid unnecessarily obscuring the concept of the invention.

[0125] The connections mentioned in this invention are divided into fixed connections and detachable connections. Fixed connections, also known as non-detachable connections, include but are not limited to conventional fixed connection methods such as folded connections, riveted connections, adhesive connections, and welded connections. Detachable connections include but are not limited to conventional disassembly methods such as bolted connections, snap-fit ​​connections, pin connections, and hinged connections. When a specific connection method is not explicitly defined, it is assumed that at least one existing connection method can be found to achieve this function, and those skilled in the art can choose according to their needs. For example, a welded connection can be chosen for fixed connections, and a bolted connection can be chosen for detachable connections.

[0126] The present invention will be further described in detail below with reference to the accompanying drawings. The following embodiments are explanations of the present invention, but the present invention is not limited to the following embodiments.

[0127] Example 1: As Figures 1-3 As shown, a longitudinal load in-situ calibration device for a wind tunnel standard force measurement model includes a standard model 1, a balance 3, a support rod 4, an angle sensor 6, and a loading rod 8. The standard model 1 is connected to the front end of the support rod 4 via the balance 3, and the rear end of the support rod 4 is connected to the roll-pitch support mechanism of the wind tunnel test section. The standard model 1 is located in the test section of the wind tunnel and is positioned facing the airflow. A model loading plate 5 is provided on the lower side of the standard model 1, and an angle sensor 6 is provided on the model loading plate 5. The upper end of the loading rod 8 is hinged to the model loading plate 5 via a universal joint 10, and several weights 9 are provided at the bottom of the loading rod 8. The standard model 1 is loaded by the weights 9.

[0128] The model loading plate 5 is set horizontally, and two T-slots 16 are provided on the lower end surface of the model loading plate 5. One of the T-slots 16 is set in the same direction as the support rod 4. The universal joint 10 includes a ball seat 11 and a loading ball 12 that are rotatably engaged. The upper end of the loading rod 8 is connected to the loading ball 12. The ball seat 11 is slidably set in any T-slot 16.

[0129] The T-slot 16 has a first scale 13 on one side, with a minimum scale value of 1mm. The ball seat 11 has a second scale 17, with a minimum scale value of 0.9mm. When the ball seat 11 slides in the T-slot 16, based on the principle of vernier calipers, the first scale 13 is used as the main scale and the second scale 17 is used as the vernier scale. By utilizing the difference in scale interval between the two, the sliding ball seat 11 can be accurately positioned.

[0130] The loading rod 8 is threadedly connected to the loading ball 12. The standard model 1 is provided with a conical hole. The front end of the balance 3 is provided with a conical head. The conical head is adapted to the conical hole and is inserted and fitted, and is tightened and fixed by the first screw 2. The model loading plate 5 is connected to the standard model 1 by the second screw 7.

[0131] Example 2: Figures 1-6 As shown, a method for in-situ calibration of longitudinal loads on a wind tunnel standard force measurement model is implemented based on the in-situ calibration device for longitudinal loads on a wind tunnel standard force measurement model described in Example 1, and includes the following steps:

[0132] Step 1: Establish the balance formula;

[0133] Step 2: Measure the distance between the two centers;

[0134] Step 3: Measure the installation angle;

[0135] Step 4: Convert the components of balance 3 to the axis system of standard model 1 to obtain the converted value of the longitudinal component;

[0136] Step 5: Apply the test load to standard model 1 to obtain the nominal value of the test load;

[0137] Step 6: Calculate the nominal deviation of the longitudinal component between the converted value and the nominal value, and determine whether the nominal deviation is acceptable. If the nominal deviation is acceptable, the calibration is complete. If the nominal deviation exceeds the tolerance, the principal coefficient is corrected, and the process returns to Step 1 to re-establish the balance formula and iterate the calibration.

[0138] The specific steps for step two are as follows:

[0139] First, remove the loading rod 8 and several weights 9. Use the angle sensor 6 to level the standard model 1. Select a loading position at a specified distance from the torque reference point of the standard model 1. This torque reference point is the inherent position of the standard model 1. Generally, the center of gravity 14 of the standard model 1 is used as the torque reference point. The specified distance is the distance given when loading the load. Make sure that the product of the loading force and the specified distance does not exceed the range. Then, install the loading rod 8 and several weights 9. Load the standard model 1 at the loading position through the loading rod 8 and several weights 9, and level the standard model 1 again. Read the measurement value of the six components of the balance 3.

[0140] Calculate the distance between the center 15 of the balance scale 3 and the torque reference point using the following formula:

[0141] ;

[0142] In the formula, Mz b The pitching moment measured by balance 3;

[0143] Y b The normal force measured by balance 3;

[0144] LP x This is the distance between the torque reference point and the axis of the loading rod 8.

[0145] The specific steps for step three are as follows:

[0146] Step 31, establish the installation angle formula:

[0147] α = α0 + α t ;

[0148] In the formula, α is the installation angle between the standard model and the balance; α0 is the fixed installation angle between standard model 1 and balance 3 when there is no load.

[0149] α t α is the change in the installation angle between the standard model 1 and the balance 3 under load, i.e., the elastic installation angle. t The linear relationship between the load and the load is as follows:

[0150] α t =k Mz Mz;

[0151] In the formula, k Mz This is a correction factor for the flexible installation angle;

[0152] Define the center of the large diameter end of the cone of balance 3 as the connection point between balance 3 and standard model 1, and Mz as the torque for converting forces and moments in the balance coordinate system to the connection point. Mz is calculated by the following formula:

[0153] Mz=Y b L+Mz b ;

[0154] In the formula, L is the lever arm of the normal force, which is calculated using the following formula:

[0155] L=L x +L0;

[0156] In the formula, L0 is the distance between the torque reference point and the connection point in the x-axis direction of the standard model 1 coordinate system;

[0157] Step 32, loading in both directions:

[0158] With the standard model 1 in an upright position, adjust it to a horizontal position using the angle sensor 6. Suspend the loading rod 8 and several weights 9 at the loading position, and record the axial force X measured by the balance 3. b+ and normal force Y b+ And calculate the installation angle α between the standard model 1 and the balance 3 in the upright state according to the following formula. + :

[0159] ;

[0160] With the standard model 1 in an inverted position (back facing down), the model loading plate 5 is moved to the back of the standard model 1. The standard model 1 is adjusted to be horizontal using the angle sensor 6. The loading rod 8 and several weights 9 are suspended at the loading position, and the axial force X measured by the balance 3 is recorded. b- and reverse normal force Y b- The installation angle α between the standard model 1 and the balance 3 in the reversed state is obtained according to the following formula. - :

[0161] ;

[0162] Step 33, separate the fixed mounting angle from the flexible mounting angle:

[0163] Taking advantage of the fact that the fixed mounting angle remains unchanged in both the upright and reverse mounting states, while the flexible mounting angle is related to the polarity of Mz, the fixed mounting angle α0 and the flexible mounting angle α are determined. t :

[0164] ;

[0165] ;

[0166] Step 34, calculate the flexible installation angle correction coefficient k Mz :

[0167] .

[0168] The specific steps for step four are as follows:

[0169] Transform the component vectors of balance 3 to the coordinate system of Standard Model 1:

[0170] Y m =Y b cos(α0+k Mz Mz);

[0171] Mz m =Mz b +Y m L x ;

[0172] X m =Y b sin(α0+k Mz Mz);

[0173] In the formula, Y m This is the value of the normal force of balance 3 converted to the axis system of standard model 1;

[0174] Mzm This is the pitch moment of balance 3 converted to the axis system value of standard model 1;

[0175] X m This is the value of the axial force of balance 3 converted to the shaft system of standard model 1.

[0176] The specific steps for step five are as follows:

[0177] The test load is applied to the standard model 1 using the model loading plate 5. The test load under the shaft system of the standard model 1 is expressed as follows:

[0178] Y N =Gcosα N ;

[0179] Mz N =Y N LP x ;

[0180] X N =Gsinα N ;

[0181] In the formula, Y N The nominal value of the normal force of the test load under the standard model 1 shaft system;

[0182] Mz N The nominal value of the pitching moment for the test load under the standard model 1 shaft system;

[0183] X N The nominal value of the axial force for the test load under the standard model 1 shaft system;

[0184] α N The nominal angle of attack for standard model 1 is measured by angle sensor 6.

[0185] G represents the nominal weight of the loading rod 8 and several weights 9.

[0186] The specific steps for step six are as follows:

[0187] The nominal deviation between the converted value and the nominal value of the corresponding component is:

[0188] ;

[0189] ;

[0190] ;

[0191] In the formula, δ X δ represents the nominal deviation between the converted value and the nominal value of the axial force component. Y δ represents the nominal deviation between the converted value and the nominal value of the normal force component. MzThis represents the nominal deviation between the upconverted value and the nominal value of the pitch moment component.

[0192] Determine δ X δ Y and δ Mz Does it exceed 120% of the accuracy range given in the balance's calibration certificate? If δ X δ Y and δ Mz If the result is within 120% accuracy of the balance 3, the calibration is complete; otherwise, correct the main term coefficient and return to step one to re-establish the balance formula and iterate the calibration.

[0193] The specific steps for correcting the principal term coefficients are as follows:

[0194] Step 61, apply the modified load to standard model 1:

[0195] Based on the required distance between the two centers and the fixed installation angle, adjust the standard model 1 so that the axis of the balance 3 is horizontal, and adjust the axial position of the ball seat 11. When the absolute value of the axial force output of the balance 3 is less than or equal to 0.01N, take the current position of the ball seat 11 as the normal force loading correction position, and apply the normal force correction load to the standard model 1.

[0196] Choose a front and rear span of L. c Two known loading points are used as pitch moment loading correction positions. First, weights 9 of equal weight are hung at the two pitch moment loading correction positions respectively. Then, the weights 9 at the two pitch moment loading correction positions are swapped to reload the pitch moment correction load.

[0197] Stand the balance upright and apply an axial force correction load;

[0198] Step 62, calculate the principal coefficient:

[0199] Calculate the principal coefficient using the following formula:

[0200] ;

[0201] In the formula, The principal coefficient of the i-th component;

[0202] This is the j-th calibration load applied to the i-th component;

[0203] To load the i-th component signal corresponding to the j-th calibration load;

[0204] Step 63, constantization of the principal term coefficients:

[0205] Offline calibration formula Undetermined coefficients in Replace with The general formula for calibrating the balance becomes:

[0206] ;

[0207] In the formula, F i The load is the i-th component of the balance.

[0208] F j The load of the j-th component that interferes with the i-th component;

[0209] F k The load of the k-th component that interferes with the i-th component;

[0210] The i-th component signal is output.

[0211] The principal coefficient of the i-th component;

[0212] The first-order interference correction coefficient of the j-th component to the i-th component;

[0213] Here are the squared interference correction factors (j=k) or cross-interference correction factors (j≠k) for each component load on the i-th component, with the following units:

[0214] The force-force interference coefficient for force components is expressed in N / N. 2 ;

[0215] The force-force disturbance coefficient for the torque component is expressed in Nm / N. 2 ;

[0216] The force-torque disturbance coefficient for force components is expressed in N / N•Nm.

[0217] The interference coefficient of the force-torque component is expressed in Nm / N•Nm.

[0218] The torque-torque disturbance coefficient of the force component is expressed in N / Nm. 2 ;

[0219] The cross-interference coefficient of torque-torque components is expressed in Nm / Nm. 2 ;

[0220] The force-force cross-term interference coefficient for the torque component is expressed in Nm / N. 2 ;

[0221] At this point, the coefficient of the principal term becomes a constant;

[0222] Step 64: Recalculate the balance formula 3 using the offline calibration data:

[0223] Calling the voltage signal output matrix during ground offline calibration [ U] and calibration load matrix [P]:

[0224] ;

[0225] ;

[0226] In the formula, m is the total number of loading groups. For a six-component balance 3, m = 144 in general.

[0227] The least squares model for calculating the balance formula is as follows:

[0228] ;

[0229] In the formula, q is the loading group number, with a total of m groups, where m is greater than the number of undetermined coefficients. The voltage signal output for the i-th element when the q-th group is loaded; The load of the i-th element when the q-th group is loaded; The load of the j-th element when the q-th group is loaded; The load of the k-th element when the q-th group is loaded;

[0230] Solve the following system of equations to obtain the undetermined coefficients. , :

[0231] ;

[0232] ;

[0233] After obtaining the new formula for balance scale 3, return to step one. If Once the calculated load accuracy reaches the nominal accuracy of the balance 3 calibration certificate, the balance 3 in-situ loading correction is complete; otherwise, continue iterative calculation.

[0234] The working principle of this invention is:

[0235] This invention discloses an in-situ calibration device and method for longitudinal load on a wind tunnel standard force measurement model. By performing in-situ loading on the wind tunnel standard force measurement model at the wind tunnel site, the balance measurement error caused by "state difference" is corrected.

[0236] Specifically, this invention designs a model loading plate 5 with a first graduation 13 and a ball seat 11 with a second graduation 17, modeled after the principle of a vernier caliper. The model loading plate 5 is rigidly connected to the standard model 1. Force is transmitted through a universal joint 10, loading is performed by a static weight 9, and monitoring is conducted by a high-precision angle sensor 6, ensuring accurate loading of the standard model 1 in the wind tunnel. By loading the standard model 1 in situ, the center distance and installation angle are obtained. Simultaneously, the fixed installation angle and the elastic installation angle are separated based on the relationship between the elastic installation angle and the load. Based on the center distance data, fixed installation calibration data, and the elastic installation angle correction formula, the components of the balance 3 are converted to the axis system of the standard model 1. By comparing the nominal deviations of the nominal values ​​and the converted values ​​under the axis system of standard model 1, the true accuracy of balance 3 is verified under the same axis system conditions. Then, based on the nominal deviation results, single-component in-situ calibration is performed on the unqualified components. The principal coefficients of the unqualified components are replaced with the principal coefficients obtained from the in-situ calibration and constantized. A new calibration formula is established, and the balance formula is recalculated by calling offline calibration data. After repeated iterative corrections, the measurement error of balance 3 meets the requirements.

[0237] The above embodiments are merely illustrative examples of the present invention and do not limit its scope of protection. Those skilled in the art can make partial changes to them, as long as they do not exceed the spirit and essence of the present invention, they are all within the scope of protection of the present invention.

Claims

1. A longitudinal load in-situ calibration device for a standard wind tunnel force measurement model, characterized in that: It includes a standard model (1), a balance (3), a support rod (4), an angle sensor (6), and a loading rod (8); the standard model (1) is connected to the support rod (4) through the balance (3), a model loading plate (5) is provided on the lower side of the standard model (1), an angle sensor (6) is provided on the model loading plate (5), the upper end of the loading rod (8) is hinged to the model loading plate (5) through a universal joint (10), and several weights (9) are provided at the bottom of the loading rod (8).

2. The in-situ calibration device for longitudinal load of a wind tunnel standard force measurement model according to claim 1, characterized in that: The model loading plate (5) is set horizontally, and two T-shaped grooves (16) are provided on the lower end surface of the model loading plate (5) and intersecting vertically. The universal joint (10) includes a ball seat (11) and a loading ball (12) that are rotated together. The upper end of the loading rod (8) is connected to the loading ball (12), and the ball seat (11) is slidably set in any T-shaped groove (16). The T-slot (16) has a first scale (13) on one side, the minimum scale value of the first scale (13) is 1mm, and the ball seat (11) has a second scale (17), the minimum scale value of the second scale (17) is 0.9mm.

3. The in-situ calibration device for longitudinal load of a wind tunnel standard force measurement model according to claim 2, characterized in that: The loading rod (8) is threadedly connected to the loading ball (12). The standard model (1) is provided with a conical hole. The front end of the balance (3) is provided with a conical head. The conical head is adapted to the conical hole and is plugged in and fixed by the first screw (2). The model loading plate (5) is connected to the standard model (1) by the second screw (7).

4. A method for in-situ calibration of longitudinal load on a standard wind tunnel force measurement model, implemented using the in-situ calibration device for longitudinal load on a standard wind tunnel force measurement model as described in claim 3, characterized in that... Includes the following steps: Step 1: Establish the balance formula; Step 2: Measure the distance between the two centers; Step 3: Measure the installation angle; Step 4: Convert the components of the balance (3) to the axis system of the standard model (1) to obtain the converted value of the longitudinal component; Step 5: Apply the test load to the standard model (1) to obtain the nominal value of the test load; Step 6: Calculate the nominal deviation of the longitudinal component between the converted value and the nominal value, and determine whether the nominal deviation is acceptable. If the nominal deviation is acceptable, the calibration is complete. If the nominal deviation exceeds the tolerance, the principal coefficient is corrected, and the process returns to Step 1 to re-establish the balance formula and iterate the calibration.

5. The method for in-situ calibration of longitudinal loads on a standard wind tunnel force measurement model according to claim 4, characterized in that, The specific steps for step two are as follows: First, remove the loading rod (8) and several weights (9), use the angle sensor (6) to level the standard model (1), select the loading position at a specified distance from the torque reference point of the standard model (1), then install the loading rod (8) and several weights (9), load the loading position of the standard model (1) through the loading rod (8) and several weights (9), and level the standard model (1) again, and read the measurement value of the six components of the balance (3); The distance between the center (15) of the balance (3) and the torque reference point is calculated according to the following formula: ； In the formula, Mz b The pitching moment measured by the balance (3); Y b The normal force measured by the balance (3); LP x The distance between the torque reference point and the axis of the loading rod (8) is denoted as .

6. The method for in-situ calibration of longitudinal loads on a standard wind tunnel force measurement model according to claim 5, characterized in that, The specific steps for step three are as follows: Step 31, establish the installation angle formula: α=α0+α t ； In the formula, α is the installation angle between the standard model (1) and the balance (3); α0 is the fixed installation angle between the standard model (1) and the balance (3) when there is no load; α t α is the change in the mounting angle between the standard model (1) and the balance (3) under load, i.e., the elastic mounting angle. t The linear relationship between the load and the load is as follows: α t =k Mz Mz; In the formula, k Mz This is a correction factor for the flexible installation angle; Define the center of the large diameter end of the cone of the balance (3) as the connection point between the balance (3) and the standard model (1), and Mz as the torque that transforms the force and torque in the balance coordinate system to the connection point. Mz is calculated by the following formula: Mz=Y b L+Mz b ; In the formula, L is the lever arm of the normal force, which is calculated using the following formula: L=L x +L0; In the formula, L0 is the distance between the torque reference point and the connection point in the x-axis direction of the standard model (1) coordinate system; Step 32, loading in both directions: With the standard model (1) in the upright position, adjust the standard model (1) to be horizontal using the angle sensor (6), suspend the loading rod (8) and several weights (9) at the loading position, and record the upright axial force X measured by the balance (3). b+ and normal force Y b+ And calculate the installation angle α between the standard model (1) and the balance (3) in the upright state according to the following formula. + : ； The standard model (1) is placed in an inverted position, with its back facing down. The model loading plate (5) is then moved to the back of the standard model (1). The standard model (1) is adjusted to a horizontal position using the angle sensor (6). The loading rod (8) and several weights (9) are suspended at the loading position. The axial force X measured by the balance (3) is recorded. b- and reverse normal force Y b- The installation angle α between the standard model (1) and the balance (3) in the reverse installation state is obtained according to the following formula. - : ； Step 33, separate the fixed mounting angle from the flexible mounting angle: Taking advantage of the fact that the fixed mounting angle remains unchanged in both the upright and reverse mounting states, while the flexible mounting angle is related to the polarity of Mz, the fixed mounting angle α0 and the flexible mounting angle α are determined. t : ； ； Step 34, calculate the flexible installation angle correction factor k Mz : 。 7. The method for in-situ calibration of longitudinal loads on a standard wind tunnel force measurement model according to claim 6, characterized in that, The specific steps for step four are as follows: Transform the component vectors of the balance (3) to the coordinate system of the standard model (1): Y m =Y b cos(α0+k Mz Mz); Mz m =Mz b +Y m L x ; X m =Y b sin(α0+k Mz Mz); In the formula, Y m The normal force of the balance (3) is converted to the axis of the standard model (1); Mz m The pitching moment of the balance (3) is converted to the axis system value of the standard model (1); X m The axial force of the balance (3) is converted to the shaft system of the standard model (1).

8. The method for in-situ calibration of longitudinal loads on a standard wind tunnel force measurement model according to claim 7, characterized in that, The specific steps for step five are as follows: The test load is applied to the standard model (1) by the model loading plate (5). The test load under the shaft system of the standard model (1) is expressed as: AND N =Gcosα N ; Mz N =Y N LP x ; X N =Gsinα N ; In the formula, Y N The nominal value of the normal force of the test load under the standard model (1) shaft system; Mz N The nominal value of the pitching moment of the test load under the standard model (1) shaft system; X N The nominal value of the axial force of the test load under the standard model (1) shaft system; α N The nominal angle of attack of the standard model (1) is measured by the angle sensor (6); G is the nominal weight of the loading rod (8) and several weights (9).

9. The method for in-situ calibration of longitudinal loads on a standard wind tunnel force measurement model according to claim 8, characterized in that, The specific steps for step six are as follows: The nominal deviation between the converted value and the nominal value of the corresponding component is: ； ； ； In the formula, δ X δ represents the nominal deviation between the converted value and the nominal value of the axial force component. Y δ represents the nominal deviation between the converted value and the nominal value of the normal force component. Mz This represents the nominal deviation between the upconverted value and the nominal value of the pitch moment component. Determine δ X δ Y and δ Mz Does it exceed 120% of the accuracy range of the balance (3)? If δ X δ Y and δ Mz If the accuracy is within 120% of the balance (3), the calibration is complete; otherwise, the main term coefficient is corrected, and the balance formula is re-established in step one for iterative calibration.

10. The method for in-situ calibration of longitudinal loads on a standard wind tunnel force measurement model according to claim 9, characterized in that, The specific steps for correcting the principal term coefficients are as follows: Step 61, apply the modified load to the standard model (1): Based on the required distance between the two centers and the fixed installation angle, adjust the standard model (1) so that the axis of the balance (3) is horizontal, adjust the axial position of the ball seat (11), and when the absolute value of the axial force output of the balance (3) is less than or equal to 0.01N, take the current position of the ball seat (11) as the normal force loading correction position, and load the normal force correction load on the standard model (1). Choose a front and rear span of L. c Two known loading points are used as pitch moment loading correction positions. First, weights of equal weight (9) are hung at the two pitch moment loading correction positions respectively. Then, the weights (9) at the two pitch moment loading correction positions are swapped to reload the pitch moment correction load. Stand the balance (3) upright and apply an axial force correction load; Step 62, calculate the principal coefficient: Calculate the principal coefficient using the following formula: ； In the formula, The principal coefficient of the i-th component; This is the j-th calibration load applied to the i-th component; To load the i-th component signal corresponding to the j-th calibration load; Step 63, constantization of the principal term coefficients: Offline calibration formula Undetermined coefficients in Replace with The general formula for calibrating the balance (3) becomes: ； In the formula, F i The load of the i-th component of the balance (3); F j The load of the j-th component that interferes with the i-th component; F k The load of the k-th component that interferes with the i-th component; The i-th component signal is output. The principal coefficient of the i-th component; The first-order interference correction coefficient of the j-th component to the i-th component; Here are the squared interference correction factors (j=k) or cross-interference correction factors (j≠k) for each component load on the i-th component, with the following units: The force-force interference coefficient for force components is expressed in N / N. 2 ; The force-force disturbance coefficient for the torque component is expressed in Nm / N. 2 ; The force-torque disturbance coefficient for force components is expressed in N / N•Nm. The interference coefficient of the force-torque component is expressed in Nm / N•Nm. The torque-torque disturbance coefficient of the force component is expressed in N / Nm. 2 ; The cross-interference coefficient of torque-torque components is expressed in Nm / Nm. 2 ; The force-force cross-term interference coefficient for the torque component is expressed in Nm / N. 2 ; At this point, the coefficient of the principal term becomes a constant; Step 64: Recalculate the balance formula (3) using the offline calibration data: Calling the voltage signal output matrix during ground offline calibration [ U] and calibration load matrix [P]: ； ； In the formula, m is the total number of loading groups. For a six-component balance (3), m = 144. The least squares model for calculating the balance (3) formula is as follows: ； In the formula, q is the loading group number, with a total of m groups, where m is greater than the number of undetermined coefficients. The voltage signal output for the i-th element when the q-th group is loaded; The load of the i-th element when the q-th group is loaded; The load of the j-th element when the q-th group is loaded; The load of the k-th element when the q-th group is loaded; Solve the following system of equations to obtain the undetermined coefficients. , : ； ； After obtaining the new formula for the balance (3), return to step one, if Once the calculated load accuracy reaches the nominal accuracy of the balance (3), the balance (3) is corrected by in-situ loading; otherwise, iterative calculation continues.

Images