A method for determining the measurement uncertainty of a wind tunnel strain gauge balance between partitions

Abstract

The present application belongs to the technical field of calibration of aerospace force test devices, and particularly relates to a method for determining wind tunnel strain balance measurement uncertainty in intervals, comprising: determining a plurality of calibration load intervals for static calibration of the balance; determining a load load implementation table and a comprehensive load error verification load implementation table for static calibration of the balance, and simultaneously loading six components on the balance to be calibrated to obtain actual output signal values of each component of the Wheatstone bridge of the balance to be calibrated and verification output signal values of each component of the Wheatstone bridge of the balance; using a response surface method, obtaining a working formula of the balance to be calibrated corresponding to the calibration load interval according to different calibration load intervals and actual output signal values of each component of the Wheatstone bridge of the balance; verifying the working formula of each balance according to the verification output signal values of each component of the Wheatstone bridge corresponding to the calibration load interval, and obtaining wind tunnel strain balance measurement uncertainty of the balance in different calibration load intervals.

Description

Technical Field

[0001] The invention belongs to the technical field of calibration of aerospace force measuring test equipment, and in particular relates to a method for determining the measurement uncertainty of a wind tunnel strain balance by intervals. Background Art

[0002] The measurement uncertainty of a wind tunnel strain gauge balance characterizes the measurement accuracy of the calibrated balance. The current assessment of wind tunnel strain gauge balance measurement uncertainty is based on GJB 2244A-2011, which stipulates that the measurement uncertainty of each balance component is affected by the following factors:

[0003] First, the uncertainty of each component of the balance introduced by the comprehensive loading repeatability u i1 ;

[0004] The second is the uncertainty u of each component of the balance introduced by the comprehensive loading error i2 ;

[0005] The third is the uncertainty of each component of the balance introduced by the uncertainty of the load source u i3 ;

[0006] Fourth, the uncertainty u of each component of the balance introduced by the data acquisition system i4 ;

[0007] The fifth is the uncertainty u of each component of the balance introduced by the calibration equipment i5 .

[0008] Therefore, the final uncertainty of the balance is calculated according to formula (1):

[0009]

[0010] Among them, u i3 、u i4 、u i5 It belongs to the "superior" equipment, and its measurement error is determined by the performance of the equipment itself. When the load source uses M1-level weights, its relative standard uncertainty is about 8×10 -6 , the measurement uncertainty of the high-precision acquisition system is about 5×10 -5 The relative standard uncertainty of each component of the calibration equipment is about 3×10 -4 , they have little effect on the balance measurement uncertainty. i1 and u i2 Related to the performance of the balance itself, the balance's comprehensive loading repeatability error is small (about 1×10 -4 ), which has little impact on the balance measurement uncertainty. Therefore, the measurement uncertainty of the balance is mainly affected by the comprehensive loading error.

[0011] The comprehensive loading error W of each component of the calibrated balance ziCalculate according to formula (2):

[0012]

[0013] Among them, P ij F represents the comprehensive loading test load value of the jth group of the i-th component of the balance; ij represents the comprehensive loading measurement value of the jth group of the i-th component of the balance; m represents the number of groups of comprehensive loading test loads, usually 15; P imax Indicates the maximum design / calibration load of each component of the balance.

[0014] From formula (2), we can see that the comprehensive loading error W of each component of the calibrated balance is zi Mainly affected by F ij and P imax Impact. ij It is related to the accuracy of the coefficients in the balance formula. The accuracy of the formula affects the comprehensive loading error W of each component of the balance. zi ; When (F ij -P ij ) takes a certain value, the comprehensive loading error W of each component of the balance zi Affected by P imax Influence.

[0015] The actual maximum aerodynamic load P that the model bears in the wind tunnel test rmax There is a load smaller than the balance design / calibration load P imax The maximum value of the balance calibration load is the balance design load P imax The coefficients in the balance formula include information about the maximum calibration load. The (non-) linear balance formula fitted based on the sample points within the calibration load range may not include the actual maximum aerodynamic load P. rmax Therefore, the maximum design / calibration load P imax The balance formula obtained by calibration has a certain error when used in wind tunnel tests. In addition, the maximum load used to calculate the measurement uncertainty when calibrating the balance is the design / calibration load P of the balance. imax The measurement uncertainty calculated by formula (2) will be less than the actual maximum aerodynamic load P rmax The measurement uncertainty of the denominator causes deviations in the calculation of the balance measurement uncertainty, affecting the assessment of the accuracy of the test data. Summary of the Invention

[0016] To address the above-mentioned deficiencies in the prior art, the present invention proposes a method for determining the measurement uncertainty of a wind tunnel strain balance by intervals. The method is used to evaluate the differential measurement uncertainty of a wind tunnel strain balance at different load intervals within the design load range. The method comprises:

[0017] According to the test range of the wind tunnel strain balance application, multiple calibration load intervals for the static calibration of the balance are determined; based on each calibration load interval, the loading load implementation table and the comprehensive loading error verification load implementation table for the static calibration of the balance are determined, and the six components of the calibrated balance are loaded simultaneously to obtain the actual output signal value of the Wheatstone bridge of each component of the calibrated balance and the verification output signal value of the Wheatstone bridge of each component of the balance; using the response surface methodology, according to different calibration load intervals and the actual output signal value of the Wheatstone bridge of each component of the balance, the working formula of the calibrated balance in the corresponding calibration load range is obtained; based on the Wheatstone bridge verification output signal value of each component of the balance in the corresponding calibration load interval, the working formula of each balance is verified, and the wind tunnel strain balance measurement uncertainty of the balance in different calibration load intervals is obtained.

[0018] As one of the improvements to the above technical solution, the method specifically includes:

[0019] Step 1) dividing the wind tunnel strain balance according to its application test range, determining multiple calibration load intervals for static calibration of the balance, and a comprehensive loading error verification load set table corresponding to each calibration load interval;

[0020] Step 2) determining a corresponding multivariate calibration load sequence table for the calibrated balance within each calibration load interval;

[0021] Step 3) converting the loading load sequence table obtained in step 2) to obtain a static calibration loading load implementation table; converting the comprehensive loading error verification load set table obtained in step 1) to obtain a comprehensive loading error verification load implementation table;

[0022] Step 4) Based on the static calibration loading load implementation table and the comprehensive loading error verification load implementation table generated in step 3), the six components of the calibrated balance are simultaneously loaded, and the actual output signal value of the Wheatstone bridge of each component of the calibrated balance and the verification output signal value of the Wheatstone bridge of each component of the calibrated balance are obtained and recorded;

[0023] Step 5) using a response surface methodology, obtaining a corresponding working formula of the calibrated balance based on different calibration load intervals and the recorded actual output signal values ​​of the Wheatstone bridge of each component of the calibrated balance;

[0024] Step 6) Substituting the Wheatstone bridge verification output signal values ​​of each component of the calibrated balance obtained in step 4) into the corresponding balance working formula obtained in step 5) for verification, and calculating the wind tunnel strain balance measurement uncertainty of the calibrated balance in different load ranges.

[0025] As one of the improvements to the above technical solution, step 1) specifically includes:

[0026] According to the test range of the wind tunnel strain balance, the eight corresponding static calibration load intervals are determined according to 20%, 40%, 60%, 80%, 100%, and 120% of the balance design load, and the balance's normal force component load Y, the balance's pitching moment component load Mz, the balance's axial force component load X, and the balance's rolling moment component load Mx are 100% or 20% of the balance design load respectively, while the balance's lateral force component load Z and the balance's yaw moment component load My are 20% or 100% of the balance design load respectively:

[0027] The plurality of static calibration load intervals include: a first load interval, a second load interval, a third load interval, a fourth load interval, a fifth load interval, a sixth load interval, a seventh load interval and an eighth load interval;

[0028] Based on the static calibration load intervals obtained above, a comprehensive loading error verification load set table corresponding to each static calibration load interval is obtained;

[0029] The comprehensive loading error verification load set table corresponding to each calibration load interval includes: historical test data or theoretical calculation data of the balance, and a predetermined number of load data randomly generated within the calibration load interval.

[0030] As one of the improvements to the above technical solution, step 2) specifically includes:

[0031] Using Design-Expert software, determine the corresponding multivariate calibration load sequence table for the calibrated balance within each calibration load range;

[0032] The sequence table includes: 6 independent variables and 6 dependent variables;

[0033] Among them, the six independent variables are the normal force component load Y of the balance, whose unit is Kg, the pitching moment component load Mz of the balance, whose unit is Kg.m, the axial force component load X of the balance, whose unit is Kg, the rolling moment component load Mx of the balance, whose unit is Kg.m, the lateral force component load Z of the balance, whose unit is Kg and the yaw moment component load My of the balance, whose unit is Kg.m;

[0034] The six dependent variables are the bridge signal output value RY of the balance normal force component, the bridge signal output value RMz of the balance pitch moment component, the bridge signal output value RX of the balance axial force component, the bridge signal output value RMx of the balance rolling moment component, the bridge signal output value RZ of the balance lateral force component and the bridge signal output value RMMy of the balance yaw moment component, and their units are all mV.

[0035] As one of the improvements of the above technical solutions, in the step 3), the loading load sequence table obtained in the step 2) is converted to obtain a static calibration loading load implementation table; the specific process includes:

[0036] According to the force system layout and the force arm size of the calibration device, the force and the moment load in the loading load sequence table are decomposed into force values required to be applied by each force applying point in the calibration device, and the multi-element calibration loading load sequence table of the calibrated scale obtained in the step 2) is converted to form a static calibration loading load implementation table.

[0037] The specific conversion process adopts the following load conversion relationship:

[0038] Y=Y1+Y2-(Y3+Y4) (3)

[0039] (Mz / L Mz )=Y1-Y2 (4)

[0040] X=X2+X4-X1-X3 (5)

[0041] (Mx / L Mx )=Y3-Y4 (6)

[0042] Z=Z1+Z2-Z3-Z4 (7)

[0043] (My / L My )= Z2+Z3-Z1-Z4 (8)

[0044] Wherein, Y is a normal force component load of the scale; Mz is a pitch moment component load of the scale; X is an axial force component load of the scale; Mx is a roll moment component load of the scale; Z is a lateral force component load of the scale; My is a yaw moment component load of the scale; L Mz is the length of the pitch moment arm; L Mx is the length of the roll moment arm; L My is the length of the yaw moment arm;

[0045] X1 is a force value applied at the position of the force applying point X1 in the calibration device;

[0046] X2 is a force value applied at the position of the force applying point X2 in the calibration device;

[0047] X3 is a force value applied at the position of the force applying point X3 in the calibration device;

[0048] X4 is a force value applied at the position of the force applying point X4 in the calibration device;

[0049] Y1 is a force value applied at the position of the force applying point Y1 in the calibration device;

[0050] Y2 is the force value applied at the force application point Y2 in the calibration device;

[0051] Y3 is the force value applied at the force application point Y3 in the calibration device;

[0052] Y4 is the force value applied at the force application point Y4 in the calibration device;

[0053] Z1 is the force value applied at the force application point Z1 in the calibration device;

[0054] Z2 is the force value applied at the force application point Z2 in the calibration device;

[0055] Z3 is the force value applied at the force application point Z3 in the calibration device;

[0056] Z4 is the force value applied at the force application point Z4 in the calibration device;

[0057] The loading load sequence table obtained in step 2) is transformed by the above conversion formula to obtain a loading load implementation table for static calibration.

[0058] As one of the improvements to the above technical solution, step 4) specifically includes:

[0059] According to the static calibration load implementation table generated in step 3), the six components of the calibrated balance are loaded simultaneously, and the actual signal output value of the Wheatstone bridge of each component of the calibrated balance is obtained and recorded using a data acquisition device;

[0060] According to the comprehensive loading error verification load implementation table generated in step 3), the six components are loaded simultaneously on the calibrated balance, and the Wheatstone bridge verification output signal value of each component of the calibrated balance is obtained and recorded using a data acquisition device.

[0061] As one of the improvements to the above technical solution, step 5) specifically includes:

[0062] Using the response surface methodology, the corresponding working formula of the calibrated balance is obtained according to the currently selected calibration load range:

[0063]

[0064] Where i = 1, 2, ..., 6; j = 1, 2, ..., 6; ΔV i represents the actual output signal value of the Wheatstone bridge of the i-th component of the calibrated balance, and F represents the applied standard load; when j≠i, It represents the first-order interference coefficient of the jth component to the ith component under the currently selected calibration load range; when j=i, Indicates the principal coefficient of the i-th component or the j-th component in the currently selected calibration load range; Fj Indicates the actual applied j-th component load value; when j = k, Indicates the second-order square interference coefficient of the jth component or the kth component on the ith component under the currently selected calibration load range; when j≠k, Indicates the cross-interference coefficient of the jth component and the kth component on the ith component under the currently selected calibration load range; F k represents the kth component load actually applied;

[0065] The above formula is a general formula. For the eight calibration load intervals, eight corresponding balance working formulas with the same form will be obtained.

[0066] As one of the improvements to the above technical solution, step 6) specifically includes:

[0067] Substitute the Wheatstone bridge verification output signal values ​​of each component of the calibrated balance obtained in step 4) into the corresponding balance working formula obtained in step 5) to obtain the corresponding comprehensive verification load value P jq ; Combined with the corresponding actual measured load value, calculate the wind tunnel strain balance measurement uncertainty W of the calibrated balance in different load ranges zj :

[0068]

[0069] Among them, P jq F represents the comprehensive loading verification load value of the jth component and the qth group of the balance; jq represents the actual measured load value of the qth group of the jth component of the balance; where q = 1, 2, ... m; m represents the number of groups of comprehensive loading test loads, usually m = 15; P jmax Indicates the maximum design / calibration load of each component of the balance;

[0070] Among them, F jq =F j or F jq =F k ;

[0071] According to the preset measurement uncertainty threshold, it is judged whether the wind tunnel strain balance measurement uncertainty of the calibrated balance in different load ranges meets the requirements;

[0072] If the measurement uncertainty of the wind tunnel strain balance of the calibrated balance in different load ranges is less than or equal to a preset measurement uncertainty threshold, it is determined that the measurement uncertainty of the wind tunnel strain balance meets the requirements;

[0073] If the wind tunnel strain balance measurement uncertainty of the calibrated balance in different load ranges is greater than a preset measurement uncertainty threshold, it is determined that the wind tunnel strain balance measurement uncertainty does not meet the requirements.

[0074] The beneficial effects of the present invention compared with the prior art are:

[0075] The method of the present invention provides different balance working formulas for different load ranges, and uses the corresponding balance working formulas to calculate and measure the corresponding wind tunnel strain balance uncertainty. The tester can select the balance working formula corresponding to the load range according to the specific wind tunnel test range (load), thereby calculating a more accurate aerodynamic load and accurately determining the error band range of the test data. BRIEF DESCRIPTION OF THE DRAWINGS

[0076] Figure 1 It is a Y / Mz load scatter plot of calibration load interval 1 of a method for determining measurement uncertainty of a wind tunnel strain balance by interval according to the present invention;

[0077] Figure 2 It is a Y / Mz load scatter plot of the full calibration interval of a method for determining the measurement uncertainty of a wind tunnel strain balance by interval according to the present invention;

[0078] Figure 3 This is a simplified structural diagram of a calibration device for determining the measurement uncertainty of a wind tunnel strain balance in a partitioned manner according to the present invention;

[0079] Figure 4 The present invention is a flow chart of a method for determining the measurement uncertainty of a wind tunnel strain balance by partitioning. DETAILED DESCRIPTION

[0080] The present invention will now be further described with reference to the accompanying drawings and examples.

[0081] The present invention provides a method for determining the measurement uncertainty of a wind tunnel strain balance by intervals, which is used to evaluate the differential measurement uncertainty of a wind tunnel strain balance in different load intervals within the design load range. The differential balance formula and measurement uncertainty within the design load range of the balance obtained by this method have passed the static loading test, and the results are highly reliable.

[0082] The method comprises: determining a plurality of calibration load intervals for static calibration of a wind tunnel strain balance according to a test range of the balance application; determining a loading load implementation table and a comprehensive loading error verification load implementation table for the static calibration of the balance based on each calibration load interval, and simultaneously loading six components on the calibrated balance to obtain actual Wheatstone bridge output signal values ​​of each component of the calibrated balance and Wheatstone bridge verification output signal values ​​of each component of the balance; adopting a response surface methodology to obtain working formulas of the calibrated balance in corresponding calibration load regions according to different calibration load intervals and actual Wheatstone bridge output signal values ​​of each component of the balance; verifying the working formula of each balance according to the Wheatstone bridge verification output signal values ​​of each component of the balance in the corresponding calibration load interval, and obtaining the wind tunnel strain balance measurement uncertainty of the balance in different calibration load intervals.

[0083] like Figure 4 As shown, the method specifically includes:

[0084] Step 1) dividing the wind tunnel strain balance according to its application test range, determining multiple calibration load intervals for static calibration of the balance, and a comprehensive loading error verification load set table corresponding to each calibration load interval;

[0085] Specifically, under normal circumstances, the maximum aerodynamic load P that the model bears in the wind tunnel test is rmax Not less than the balance design load P imax 20% (adjustment that can be increased or decreased, at least any positive integer greater than 0 can be used as an adjustable value) or not greater than the balance design load P imax 120% (can be adjusted to increase or decrease, maximum not exceeding 130%); there are also cases where the longitudinal (lift direction) load is large (small) and the transverse (side direction) load is small (large);

[0086] According to the test range of the wind tunnel strain balance, the eight corresponding static calibration load intervals are determined according to 20%, 40%, 60%, 80%, 100%, and 120% of the balance design load, and the balance's normal force component load Y, the balance's pitching moment component load Mz, the balance's axial force component load X, and the balance's rolling moment component load Mx are each 100% (20%) of the balance design load. At the same time, the balance's lateral force component load Z and the balance's yaw moment component load My are each 20% (100%) of the balance design load. The above proportions can be adjusted according to the actual application of the balance to determine the following eight static calibration load intervals:

[0087] The plurality of static calibration load intervals include: a first load interval, a second load interval, a third load interval, a fourth load interval, a fifth load interval, a sixth load interval, a seventh load interval and an eighth load interval;

[0088] Based on the static calibration load intervals obtained above, a comprehensive loading error verification load set table corresponding to each static calibration load interval is obtained;

[0089] The comprehensive loading error verification load set table corresponding to each calibration load interval includes: historical test data or theoretical calculation data of the balance, and a predetermined number of load data randomly generated within the calibration load interval;

[0090] The historical test data or theoretical calculation data of the balance are classified using a clustering algorithm to obtain corresponding classification data, which are stored in corresponding load intervals.

[0091] Step 2) determining a corresponding multivariate calibration load sequence table for the calibrated balance within each calibration load interval based on experimental design method theory;

[0092] Specifically, according to the experimental design method theory, Design-Expert software is used to determine the corresponding multivariate calibration load sequence table of the calibrated balance within each calibration load range;

[0093] The sequence table includes: 6 independent variables and 6 dependent variables;

[0094] Among them, the six independent variables are the normal force component load Y of the balance, whose unit is Kg, the pitching moment component load Mz of the balance, whose unit is Kg.m, the axial force component load X of the balance, whose unit is Kg, the rolling moment component load Mx of the balance, whose unit is Kg.m, the lateral force component load Z of the balance, whose unit is Kg and the yaw moment component load My of the balance, whose unit is Kg.m;

[0095] The six dependent variables are the bridge signal output value RY of the balance normal force component, the bridge signal output value RMz of the balance pitch moment component, the bridge signal output value RX of the balance axial force component, the bridge signal output value RMx of the balance rolling moment component, the bridge signal output value RZ of the balance lateral force component and the bridge signal output value RMMy of the balance yaw moment component, and their units are all mV.

[0096] Step 3) According to the characteristics of the calibration device, the loading load sequence table obtained in step 2) is transformed to obtain a loading load implementation table for static calibration; the comprehensive loading error verification load set table obtained in step 1) is transformed to obtain a comprehensive loading error verification load implementation table;

[0097] Specifically, according to the structure of the calibration device, the multivariate calibration load sequence table of the calibrated balance obtained in step 2) is converted into a static calibration load implementation table;

[0098] Specifically, based on the force system layout and lever arm size of the calibration device, the force and torque loads in the above loading load sequence table are decomposed into the force values ​​required to be applied at each force application point in the calibration device. The multivariate calibration loading load sequence table of the calibrated balance obtained in step 2) is converted to form a static calibration loading load implementation table;

[0099] The specific conversion process is carried out using the following load conversion relationship:

[0100] Y=Y1+Y2-(Y3+Y4) (3)

[0101] (Mz / (L Mz ))=Y1-Y2 (4)

[0102] X=X2+X4-X1-X3 (5)

[0103] (Mx / (L Mx ))=Y3-Y4 (6)

[0104] Z=Z1+Z2-Z3-Z4 (7)

[0105] (My / (L My ))= Z2+Z3-Z1-Z4 (8)

[0106] Among them, Y is the normal force component load of the balance; Mz is the pitch moment component load of the balance; X is the axial force component load of the balance; Mx is the rolling moment component load of the balance; Z is the lateral force component load of the balance; My is the yaw moment component load of the balance; L Mz is the length of the arm of the pitching moment; L Mx is the length of the arm of the rolling moment; L My is the length of the lever arm of the yaw moment;

[0107] X1 is the calibration device ( Figure 3 ) the force value applied at the force application point X1;

[0108] X2 is the calibration device ( Figure 3 ) the force value applied at the force application point X2;

[0109] X3 is the calibration device ( Figure 3 ) the force value applied at the force application point X3;

[0110] X4 is the calibration device ( Figure 3 ) the force value applied at the force application point X4;

[0111] Y1 is the calibration device ( Figure 3 ) the force value applied at the force application point Y1;

[0112] Y2 is the calibration device (Figure 3 ) the force value applied at the force application point Y2;

[0113] Y3 is the calibration device ( Figure 3 ) the force value applied at the force application point Y3;

[0114] Y4 is the calibration device ( Figure 3 ) the force value applied at the force application point Y4;

[0115] Z1 is the calibration device ( Figure 3 ) the force value applied at the force application point Z1;

[0116] Z2 is the calibration device ( Figure 3 ) the force value applied at the force application point Z2;

[0117] Z3 is the calibration device ( Figure 3 ) the force value applied at the force application point Z3;

[0118] Z4 is the calibration device ( Figure 3 ) the force value applied at the force application point Z4;

[0119] The calibration device is a well-known calibration device in the art, and its structure is also well-known.

[0120] The loading load sequence table obtained in step 2) is transformed by the above conversion formula to obtain a loading load implementation table for static calibration; in other specific embodiments, other force system layouts and force arm sizes of the calibration device can also be used to perform corresponding conversions.

[0121] Similarly, the above-mentioned conversion relationship is used to convert the comprehensive loading error verification load set table obtained in step 1) into a comprehensive loading error verification load implementation table.

[0122] Step 4) Based on the static calibration loading load implementation table and the comprehensive loading error verification load implementation table generated in step 3), the six components of the calibrated balance are simultaneously loaded, and the actual output signal value of the Wheatstone bridge of each component of the calibrated balance and the verification output signal value of the Wheatstone bridge of each component of the calibrated balance are obtained and recorded;

[0123] Specifically, according to the static calibration load implementation table generated in step 3), the six components of the calibrated balance are loaded simultaneously, and the actual signal output value of the Wheatstone bridge of each component of the calibrated balance is obtained and recorded using a data acquisition device;

[0124] Specifically, the force value at each force application point of the calibration device is generated by a standard weight or a high-precision force generator, and the force value at each force application point is applied to each component of the calibrated balance. After the bridge signal output value of each component of the calibrated balance is stable (the fluctuation value of each component signal is less than 0.002mV), the data acquisition equipment records the Wheatstone bridge signal output value of each component of the calibrated balance.

[0125] According to the comprehensive loading error verification load implementation table generated in step 3), the six components are loaded simultaneously on the calibrated balance, and the Wheatstone bridge verification output signal value of each component of the calibrated balance is obtained and recorded using a data acquisition device.

[0126] Step 5) processing the data obtained in step 4) using a response surface methodology, and obtaining a corresponding working formula for the calibrated balance based on different calibration load intervals and the recorded actual output signal values ​​of the Wheatstone bridge of each component of the calibrated balance, wherein the comprehensive error verification load and its corresponding bridge signal output value are not involved in the data processing;

[0127] Specifically, the response surface methodology is used to obtain the corresponding working formula of the calibrated balance according to the current calibration load range:

[0128]

[0129] Where i = 1, 2, ..., 6; j = 1, 2, ..., 6; ΔV i represents the actual output signal value of the Wheatstone bridge of the i-th component of the calibrated balance, and F represents the applied standard load; when j≠i, It represents the first-order interference coefficient of the jth component to the ith component under the currently selected calibration load range; when j=i, Indicates the principal coefficient of the i-th component or the j-th component in the currently selected calibration load range; F j Indicates the actual applied j-th component load value; when j = k, Indicates the second-order square interference coefficient of the jth component or the kth component on the ith component under the currently selected calibration load range; when j≠k, Indicates the cross-interference coefficient of the jth component and the kth component on the ith component under the currently selected calibration load range; F k represents the kth component load actually applied;

[0130] The above formula is a general formula. For eight calibration load intervals, eight corresponding balance working formulas with the same form will be obtained. and It is determined according to the corresponding calibration load interval, that is, the expression of the balance working formula corresponding to each different calibration load interval is the same. The only difference is: and The value of is determined by the currently selected calibration load range and is different. Among them, the comprehensive error verification load and its corresponding bridge signal output value are not involved in data processing.

[0131] Step 6) Substitute the Wheatstone bridge verification output signal values ​​of each component of the calibrated balance obtained in step 4) into the corresponding balance working formula obtained in step 5) to verify and test its accuracy, and calculate the wind tunnel strain balance measurement uncertainty of the calibrated balance in different load ranges.

[0132] Specifically, the Wheatstone bridge verification output signal values ​​of each component of the calibrated balance obtained in step 4) are substituted into the corresponding balance working formula obtained in step 5) to obtain the corresponding comprehensive verification load value P jq ; Combined with the corresponding actual measured load value, calculate the wind tunnel strain balance measurement uncertainty W of the calibrated balance in different load ranges zj :

[0133]

[0134] Among them, P jq F represents the comprehensive loading verification load value of the jth component and the qth group of the balance; jq represents the actual measured load value of the qth group of the jth component of the balance; where q = 1, 2, ... m; m represents the number of groups of comprehensive loading test loads, usually m = 15; P jmax Indicates the maximum design / calibration load of each component of the balance;

[0135] Among them, F jq is F in the above balance working formula k or F j ; F jq =F j or F jq =F k ;

[0136] According to the preset measurement uncertainty threshold, it is judged whether the wind tunnel strain balance measurement uncertainty of the calibrated balance in different load ranges meets the requirements;

[0137] If the measurement uncertainty of the wind tunnel strain balance of the calibrated balance in different load ranges is less than or equal to a preset measurement uncertainty threshold, it is determined that the measurement uncertainty of the wind tunnel strain balance meets the requirements;

[0138] If the wind tunnel strain balance measurement uncertainty of the calibrated balance in different load ranges is greater than a preset measurement uncertainty threshold, it is determined that the wind tunnel strain balance measurement uncertainty does not meet the requirements.

[0139] Example 1.

[0140] The present invention provides a method for determining the measurement uncertainty of a wind tunnel strain balance by partitioning the intervals, the method specifically comprising:

[0141] Step 1) dividing the wind tunnel strain balance according to its application test range, determining multiple calibration load intervals for static calibration of the balance, and a comprehensive loading error verification load set table corresponding to each calibration load interval;

[0142] Specifically, in this example, the design load range of each component of the balance is shown in Table 1. The actual maximum aerodynamic load P borne by the wind tunnel test model used by the balance is rmax Equal to the balance design load P jmax ; The X component load in the balance is greater than 0, and the other component loads are positive or negative.

[0143] Table 1 Design load range of each component of the balance

[0144] Y (Kg) Mz (Kg.m) X (Kg) Mx (Kg.m) Z (Kg) My (Kg.m) -100～100 -32～32 0～100 -13～13 -80～80 -20～20

[0145] The design load range of each component of the balance is used as a test range for application of the wind tunnel strain balance; wherein the multiple static calibration load intervals include: a first load interval, a second load interval, a third load interval, a fourth load interval, a fifth load interval, a sixth load interval, a seventh load interval, and an eighth load interval;

[0146] According to the balance design load P imax The ranges of each component of the eight load intervals (i.e., calibration load interval 1, calibration load interval 2, calibration load interval 3, calibration load interval 4, calibration load interval 5, calibration load interval 6, calibration load interval 7, and calibration load interval 8 in Table 2) determined by 20%, 40%, 60%, 80%, 100%, and 120% of the design load, as well as the Y, Mz, X, and Mx components at 100% (20%) of the design load and the Z and My components at 20% (100%) of the design load are shown in Table 2 below, that is, each ratio corresponds to a calibration load interval.

[0147] Table 2 Static calibration load range of balance

[0148] Y (Kg) Mz (Kg.m) X (Kg) Mx (Kg.m) Z (Kg) My (Kg.m) Calibration load interval 1 -120～120 -38.4～38.4 0～120 -16～16 -96～96 -24～24 Calibration load interval 2 -100～100 -32～32 0～100 -13～13 -80～80 -20～20 Calibration load interval 3 -80～80 -25.6～25.6 0～80 -11～11 -64～64 -16～16 Calibration load interval 4 -60～60 -19.2～19.2 0～60 -8～8 -48～48 -12～12 Calibration load interval 5 -40～40 -12.8～12.8 0～40 -5.2～5.2 -32～32 -8～8 Calibration load interval 6 -20～20 -6.4～6.4 0～20 -2.6～2.6 -16～16 -4～4 Calibration load interval 7 -100～100 -32～32 0～100 -13～13 -16～16 -4～4 Calibration load interval 8 -20～20 -6.4～6.4 0～100 -13～13 -80～80 -20～20

[0149] The comprehensive loading error verification load set table corresponding to each calibration load interval consists of two parts: one is the historical test data or theoretical calculation data of the balance, and the other is a preset number of load data randomly generated within each calibration load interval.

[0150] Historical test data or theoretical calculation data are classified using a clustering algorithm and stored in corresponding load intervals. The randomly generated load data within each calibration load interval cannot be identical to the data generated by the clustering algorithm. Table 3 shows the 15 sets of comprehensive loading error verification load sets corresponding to calibration load interval 1.

[0151] Table 3 Comprehensive loading error verification load set corresponding to calibration load interval 1

[0152]

[0153]

[0154] Based on the 15 groups of comprehensive loading error verification load set tables corresponding to the calibration load interval 1 obtained above, the same method is used to obtain the comprehensive loading error verification random load set tables corresponding to the calibration load interval 2, calibration load interval 3, calibration load interval 4, calibration load interval 5, calibration load interval 6, calibration load interval 7, and calibration load interval 8 respectively.

[0155] Step 2) Based on experimental design theory, a corresponding multivariate calibration load sequence table for the calibrated balance is determined within each calibration load interval; wherein each calibration load interval includes multiple load sequences, and each recorded sequence corresponds to a multivariate calibration load for the calibrated balance;

[0156] Specifically, according to the experimental design method theory, Design-Expert software is used to determine the corresponding multivariate calibration load sequence table of the calibrated balance within each calibration load range;

[0157] The sequence table includes: 6 independent variables and 6 dependent variables;

[0158] Among them, the six independent variables are the normal force component load Y of the balance, whose unit is Kg, the pitching moment component load Mz of the balance, whose unit is Kg.m, the axial force component load X of the balance, whose unit is Kg, the rolling moment component load Mx of the balance, whose unit is Kg.m, the lateral force component load Z of the balance, whose unit is Kg and the yaw moment component load My of the balance, whose unit is Kg.m;

[0159] The six dependent variables are the bridge signal output value RY of the balance normal force component, the bridge signal output value RMz of the balance pitch moment component, the bridge signal output value RX of the balance axial force component, the bridge signal output value RMx of the balance rolling moment component, the bridge signal output value RZ of the balance lateral force component and the bridge signal output value RMMy of the balance yaw moment component, and their units are all mV.

[0160] In this example, the Box-Behnken function in the Design-Expert software is used to design each calibration load interval to generate the corresponding multivariate calibration load sequence table (see Table 4):

[0161] Table 4 Multivariate calibration load sequence corresponding to different calibration load intervals

[0162]

[0163] Each calibration load interval contains 54 loading sequences, and the 8 calibration intervals contain a total of 432 loading sequences. Each loading sequence corresponds to the multivariate calibration load of the calibrated balance; where multivariate refers to the multiple component loads of the calibrated balance. Among them, the Y / Mz component load scatter plot of calibration load interval 1 is as follows: Figure 1 As shown in the figure, the Y / Mz component load scatter diagram of the entire calibration load range (1 to 8) is as follows: Figure 2 As shown;

[0164] Figure 1 The combined relationship of the two component loads Y and Mz in the calibration load interval 1 and the position of the combined relationship in the plane coordinate system are shown;

[0165] Figure 2 Figure 1 The combined relationship of the two component loads Y and Mz in the calibration load range 1 to 8 and the position of the combined relationship in the plane coordinate system are shown;

[0166] Its purpose is to be able to visually show the relationship between the loads on each two of the six components of the balance.

[0167] Step 3) generating a static calibration loading load implementation table and a comprehensive loading error verification load implementation table according to the calibration device structure;

[0168] Specifically, according to the structure of the calibration device, the loading load sequence table obtained in step 2) is transformed to obtain a loading load implementation table for static calibration;

[0169] Specifically, according to Figure 3The force system layout and lever arm size of the calibration device shown in the figure are used to decompose the force and torque loads in the above loading load sequence table into the force values ​​that need to be applied at each force application point in the calibration device. The loading load sequence table obtained in step 2) is converted to form a loading load implementation table for static calibration.

[0170] In this example, the schematic diagram of the calibration device is shown in Figure 3 , the specific conversion process is carried out using the following load conversion relationship according to the following formula:

[0171] Y=Y1+Y2-(Y3+Y4) (3)

[0172] (Mz / (L Mz ))=Y1-Y2 (4)

[0173] X=X2+X4-X1-X3 (5)

[0174] (Mx / (L Mx ))=Y3-Y4 (6)

[0175] Z=Z1+Z2-Z3-Z4 (7)

[0176] (My / (L My ))= Z2+Z3-Z1-Z4 (8)

[0177] The loading load sequence table obtained in step 2) is converted using the above conversion formula to obtain a loading load implementation table for static calibration;

[0178] Similarly, the comprehensive loading error verification load set table obtained in step 1) may also be transformed using the above-mentioned transformation process to obtain a comprehensive loading error verification load implementation table.

[0179] Step 4) Based on the static calibration loading load implementation table and the comprehensive loading error verification load implementation table generated in step 3), the six components of the calibrated balance are simultaneously loaded, and the actual output signal value of the Wheatstone bridge of each component of the calibrated balance and the verification output signal value of the Wheatstone bridge of each component of the calibrated balance are obtained and recorded;

[0180] Specifically, according to the static calibration load implementation table generated in step 3), the six components of the calibrated balance are loaded simultaneously, and the actual signal output value of the Wheatstone bridge of each component of the calibrated balance is obtained and recorded using a data acquisition device;

[0181] Specifically, the force value at each force application point of the calibration device is generated by a standard weight or a high-precision force generator, and the force value at each force application point is applied to each component of the calibrated balance. After the bridge signal output value of each component of the calibrated balance is stable (the fluctuation value of each component signal is less than 0.002mV), the data acquisition equipment records the Wheatstone bridge signal output value of each component of the calibrated balance.

[0182] According to the comprehensive loading error verification load implementation table generated in step 3), the six components are loaded simultaneously on the calibrated balance, and the Wheatstone bridge verification output signal value of each component of the calibrated balance is obtained and recorded using a data acquisition device.

[0183] In this example, the six components of the calibrated balance are simultaneously loaded according to the load application table, with the force generated by standard weights. The bridge signal output values ​​for each component of the calibrated balance are recorded and stored by a data acquisition device. Each bridge signal output value corresponds to a load group in the load application table.

[0184] Step 5) processing the data obtained in step 4) using a response surface methodology, and obtaining a corresponding working formula for the calibrated balance based on different calibration load intervals and the recorded actual output signal values ​​of the Wheatstone bridge of each component of the calibrated balance, wherein the comprehensive error verification load and its corresponding bridge signal output value are not involved in the data processing;

[0185] Specifically, the response surface methodology is used to obtain the corresponding working formula of the calibrated balance according to the current calibration load range:

[0186]

[0187] Where i = 1, 2, ..., 6; j = 1, 2, ..., 6; ΔV i represents the actual output signal value of the Wheatstone bridge of the i-th component of the calibrated balance, and F represents the applied standard load; when j≠i, It represents the first-order interference coefficient of the jth component to the ith component under the currently selected calibration load range; when j=i, Indicates the principal coefficient of the jth component or the ith component in the currently selected calibration load range; F j Indicates the actual applied j-th component load value; when j = k, Indicates the second-order square interference coefficient of the jth component or the kth component on the ith component under the currently selected calibration load range; when j≠k, Indicates the cross-interference coefficient of the jth component and the kth component on the ith component under the currently selected calibration load range; F k represents the kth component load actually applied;

[0188] The above formula is a general formula. For eight calibration load intervals, eight corresponding balance working formulas with the same form will be obtained. and It is determined according to the corresponding calibration load interval, that is, the expression of the balance working formula corresponding to each different calibration load interval is the same. The only difference is: and The value of is determined by the currently selected calibration load range and is different. Among them, the comprehensive error verification load and its corresponding bridge signal output value are not involved in data processing.

[0189] Step 6) Substitute the Wheatstone bridge verification output signal values ​​of each component of the calibrated balance obtained in step 4) into the corresponding balance working formula obtained in step 5) to verify and test its accuracy, and calculate the wind tunnel strain balance measurement uncertainty of the calibrated balance in different load ranges.

[0190] Specifically, the Wheatstone bridge verification output signal values ​​of each component of the calibrated balance obtained in step 4) are substituted into the corresponding balance working formula obtained in step 5) to obtain the corresponding comprehensive verification load value P jq ; Combined with the corresponding actual measured load value, calculate the wind tunnel strain balance measurement uncertainty W of the calibrated balance in different load ranges zj :

[0191]

[0192] Among them, P jq F represents the comprehensive loading verification load value of the jth component and the qth group of the balance; jq represents the actual measured load value of the qth group of the jth component of the balance; where q = 1, 2, ... m; m represents the number of groups of comprehensive loading test loads, usually m = 15; P jmax Indicates the maximum design / calibration load of each component of the balance;

[0193] Among them, F jq is F in the above balance working formula k or F j ; F jq =F j or F jq =F k ;

[0194] According to the preset measurement uncertainty threshold, it is judged whether the wind tunnel strain balance measurement uncertainty of the calibrated balance in different load ranges meets the requirements;

[0195] If the wind tunnel strain gauge measurement uncertainty of the calibrated balance in different load intervals is less than or equal to the preset measurement uncertainty threshold, it is determined that the wind tunnel strain gauge measurement uncertainty is qualified;

[0196] If the wind tunnel strain gauge measurement uncertainty of the calibrated balance in different load intervals is greater than the preset measurement uncertainty threshold, it is determined that the wind tunnel strain gauge measurement uncertainty is unqualified.

[0197] In the present example, the electrical signal values corresponding to the 15 groups of comprehensive loading error verification loads of each calibration load interval obtained in step 4) are respectively brought into the corresponding balance working formula, to obtain the calculated load, and finally the measurement uncertainty of the calibrated balance in each calibration load interval is calculated by using formula (2).

[0198] In the present example, the test range (load range) of the balance is divided into multiple calibration load intervals, and in each calibration load interval, the mature Design-Expert test design software is used to design a multivariate calibration loading load sequence table, which ensures the scientificity of the load acting on the calibrated balance, and at the same time, the working efficiency of the multiple calibration load intervals is taken into account. The implementation of the calibration load interval calibration and the interval given measurement uncertainty of the balance is beneficial to improve the quality of the wind tunnel test data, and is more beneficial to the test personnel to determine the accuracy of the test data through the different measurement uncertainties of the balance.

[0199] Finally, it should be noted that the above examples are only used to illustrate the technical solutions of the present application and are not limiting. Although the present application has been described in detail with reference to the examples, those skilled in the art should understand that modifications or equivalent replacements of the technical solutions of the present application do not deviate from the spirit and scope of the present application, and they should be covered in the scope of the claims of the present application.

Claims

1. A method for determining the measurement uncertainty of a wind tunnel strain balance by intervals, used to evaluate the differential measurement uncertainty of a wind tunnel strain balance at different load intervals within its design load range; the method comprises: Step 1) dividing the wind tunnel strain balance according to its application test range, determining multiple calibration load intervals for static calibration of the balance, and a comprehensive loading error verification load set table corresponding to each calibration load interval; Step 2) determining a corresponding multivariate calibration load sequence table for the calibrated balance within each calibration load interval; Step 3) converting the loading load sequence table obtained in step 2) to obtain a static calibration loading load implementation table; converting the comprehensive loading error verification load set table obtained in step 1) to obtain a comprehensive loading error verification load implementation table; Step 4) Based on the static calibration loading load implementation table and the comprehensive loading error verification load implementation table generated in step 3), the six components of the calibrated balance are simultaneously loaded, and the actual output signal value of the Wheatstone bridge of each component of the calibrated balance and the verification output signal value of the Wheatstone bridge of each component of the calibrated balance are obtained and recorded; Step 5) using a response surface methodology, obtaining a corresponding working formula of the calibrated balance based on different calibration load intervals and the recorded actual output signal values ​​of the Wheatstone bridge of each component of the calibrated balance; Step 6) Substituting the Wheatstone bridge verification output signal values ​​of each component of the calibrated balance obtained in step 4) into the corresponding balance working formula obtained in step 5) for verification, and calculating the wind tunnel strain balance measurement uncertainty of the calibrated balance in different load ranges; The step 5) specifically includes: Using the response surface methodology, the corresponding working formula of the calibrated balance is obtained according to the currently selected calibration load range: Where i = 1, 2, ..., 6; j = 1, 2, ..., 6; ΔV i represents the actual output signal value of the Wheatstone bridge of the i-th component of the calibrated balance, and F represents the applied standard load; when j≠i, It represents the first-order interference coefficient of the jth component to the ith component under the currently selected calibration load range; when j=i, Indicates the principal coefficient of the i-th component or the j-th component in the currently selected calibration load range; F j Indicates the actual applied j-th component load value; when j = k, Indicates the second-order square interference coefficient of the jth component or the kth component on the ith component under the currently selected calibration load range; when j≠k, Indicates the cross-interference coefficient of the jth component and the kth component on the ith component under the currently selected calibration load range; F k represents the kth component load actually applied; The above formula is for eight calibration load intervals, and eight corresponding balance working formulas of the same form will be obtained.

2. The method for determining the measurement uncertainty of a wind tunnel strain balance by intervals according to claim 1, characterized in that: The step 1) specifically includes: According to the test range of the wind tunnel strain balance, the eight corresponding static calibration load intervals are determined according to 20%, 40%, 60%, 80%, 100%, and 120% of the balance design load, and the balance's normal force component load Y, the balance's pitching moment component load Mz, the balance's axial force component load X, and the balance's rolling moment component load Mx are 100% or 20% of the balance design load respectively, while the balance's lateral force component load Z and the balance's yaw moment component load My are 20% or 100% of the balance design load respectively: The plurality of static calibration load intervals include: a first load interval, a second load interval, a third load interval, a fourth load interval, a fifth load interval, a sixth load interval, a seventh load interval and an eighth load interval; Based on the static calibration load intervals obtained above, a comprehensive loading error verification load set table corresponding to each static calibration load interval is obtained; The comprehensive loading error verification load set table corresponding to each calibration load interval includes: historical test data or theoretical calculation data of the balance, and a predetermined number of load data randomly generated within the calibration load interval.

3. The method for determining the measurement uncertainty of a wind tunnel strain balance by intervals according to claim 1, characterized in that: The step 2) specifically includes: Using Design-Expert software, determine the corresponding multivariate calibration load sequence table for the calibrated balance within each calibration load range; The sequence table includes: 6 independent variables and 6 dependent variables; Among them, the six independent variables are the normal force component load Y of the balance, whose unit is Kg, the pitching moment component load Mz of the balance, whose unit is Kg.m, the axial force component load X of the balance, whose unit is Kg, the rolling moment component load Mx of the balance, whose unit is Kg.m, the lateral force component load Z of the balance, whose unit is Kg and the yaw moment component load My of the balance, whose unit is Kg.m; The six dependent variables are the bridge signal output value RY of the balance normal force component, the bridge signal output value RMz of the balance pitch moment component, the bridge signal output value RX of the balance axial force component, the bridge signal output value RMx of the balance rolling moment component, the bridge signal output value RZ of the balance lateral force component and the bridge signal output value RMMy of the balance yaw moment component, and their units are all mV.

4. The method for determining the measurement uncertainty of a wind tunnel strain balance by intervals according to claim 1, characterized in that: In the step 3), the loading load sequence table obtained in the step 2) is transformed to obtain a loading load implementation table for static calibration; The specific process includes: Based on the force system layout and lever arm size of the calibration device, the force and torque loads in the above loading load sequence table are decomposed into the force values ​​required to be applied at each force application point in the calibration device. The multivariate calibration loading load sequence table of the calibrated balance obtained in step 2) is converted to form a static calibration loading load implementation table; The specific conversion process is carried out using the following load conversion relationship: Y=Y1+Y2-(Y3+Y4) (3) (Mz / L Mz )=Y1-Y2 (4) X=X2+X4-X1-X3 (5) (Mx / L Mx )=Y3-Y4 (6) Z=Z1+Z2-Z3-Z4 (7) (My / L My )=Z2+Z3-Z1-Z4 (8) Among them, Y is the normal force component load of the balance; Mz is the pitch moment component load of the balance; X is the axial force component load of the balance; Mx is the rolling moment component load of the balance; Z is the lateral force component load of the balance; My is the yaw moment component load of the balance; L Mz is the length of the arm of the pitching moment; L Mx is the length of the arm of the rolling moment; L My is the length of the lever arm of the yaw moment; X1 is the force value applied at the force application point X1 in the calibration device; X2 is the force value applied at the force application point X2 in the calibration device; X3 is the force value applied at the force application point X3 in the calibration device; X4 is the force value applied at the force application point X4 in the calibration device; Y1 is the force value applied at the force application point Y1 in the calibration device; Y2 is the force value applied at the force application point Y2 in the calibration device; Y3 is the force value applied at the force application point Y3 in the calibration device; Y4 is the force value applied at the force application point Y4 in the calibration device; Z1 is the force value applied at the force application point Z1 in the calibration device; Z2 is the force value applied at the force application point Z2 in the calibration device; Z3 is the force value applied at the force application point Z3 in the calibration device; Z4 is the force value applied at the force application point Z4 in the calibration device; The loading load sequence table obtained in step 2) is transformed using the above load conversion relationship to obtain a loading load implementation table for static calibration.

5. The method for determining the measurement uncertainty of a wind tunnel strain balance by intervals according to claim 1, characterized in that: The step 4) specifically includes: According to the static calibration load implementation table generated in step 3), the six components of the calibrated balance are loaded simultaneously, and the actual signal output value of the Wheatstone bridge of each component of the calibrated balance is obtained and recorded using a data acquisition device; According to the comprehensive loading error verification load implementation table generated in step 3), the six components are loaded simultaneously on the calibrated balance, and the Wheatstone bridge verification output signal value of each component of the calibrated balance is obtained and recorded using a data acquisition device.

6. The method for determining the measurement uncertainty of a wind tunnel strain balance by intervals according to claim 1, characterized in that: The step 6) specifically includes: Substitute the Wheatstone bridge verification output signal values ​​of each component of the calibrated balance obtained in step 4) into the corresponding balance working formula obtained in step 5) to obtain the corresponding comprehensive verification load value P jq ; Combined with the corresponding actual measured load value, calculate the wind tunnel strain balance measurement uncertainty W of the calibrated balance in different load ranges zj : Among them, P jq F represents the comprehensive loading verification load value of the jth component and the qth group of the balance; jq represents the actual measured load value of the qth group of the jth component of the balance; where q = 1, 2, ... m; m represents the number of groups of comprehensive loading test loads, and m = 15; P jmax Indicates the maximum design / calibration load of each component of the balance; Among them, F jq =F j or F jq =F k ; According to the preset measurement uncertainty threshold, it is judged whether the wind tunnel strain balance measurement uncertainty of the calibrated balance in different load ranges meets the requirements; If the measurement uncertainty of the wind tunnel strain balance of the calibrated balance in different load ranges is less than or equal to a preset measurement uncertainty threshold, it is determined that the measurement uncertainty of the wind tunnel strain balance meets the requirements; If the wind tunnel strain balance measurement uncertainty of the calibrated balance in different load ranges is greater than a preset measurement uncertainty threshold, it is determined that the wind tunnel strain balance measurement uncertainty does not meet the requirements.

Images