Dynamic performance calibration method for rotor strain balance

Abstract

The invention provides a dynamic performance calibration method for a rotor strain balance, and the method achieves the accurate measurement of a balance for a dynamic AC force through the measurement of a transfer function and the design of a filter on the basis of the calibration of a balance loading weight, thereby enabling the measurement of a rotor pulling force through the balance during a helicopter rotor test to be more accurate.

Description

Technical Field

[0001] The present application belongs to the technical field of balance calibration, and specifically relates to a method for calibrating the dynamic performance of a rotor strain balance. Background Art

[0002] When conducting various tests on the rotating state of the helicopter rotor, the rotor is installed on a special test bench for testing. The test bench can rotate the rotor at a set speed, and the test bench has a control system that can change the total pitch and cyclic pitch of the rotor. In addition, the test bench is also equipped with a multi-component strain balance, which measures the tension and lateral force generated when the rotor rotates.

[0003] The balance needs to be calibrated before use to determine the conversion relationship between the tension and the voltage value output by the sensor on the balance. The current balance calibration method is to use weights for loading. The specific method is: within the measurement range of the balance, it is generally divided into 5 or more gears, and weights of different weights are loaded respectively. The sensor output of the balance after the weights are loaded is measured, and then the conversion relationship is obtained through calculation to complete the balance calibration. If the force borne by the balance is static or the frequency is very low, this method is completely applicable, but in fact, the tension borne by the balance during use comes from the rotating rotor, so there is a periodic component in the tension, and its change period is related to the rotor speed and the number of blades, and its frequency is usually in the range of more than ten hertz to tens of hertz. For these periodic components, the conversion relationship obtained by static calibration of weights is still used to calculate the tension, which generally leads to measurement errors. Summary of the invention

[0004] Purpose of the invention: In order to overcome the shortcomings of static weight calibration, the present invention proposes a dynamic calibration method that can accurately measure the dynamic tension borne by the balance.

[0005] The present application provides a rotor strain balance dynamic performance calibration method, the method comprising the following steps:

[0006] On the basis of the obtained transfer function, taking the DC amplitude of the transfer function as a reference, the amplitudes at other frequencies are compared with the DC amplitude to obtain a set of reference values ​​of the transfer function;

[0007] According to the obtained transfer function reference value, the reciprocal of the reference value is taken at each frequency to obtain a set of values, which are called amplitude correction coefficients at different frequencies;

[0008] The obtained amplitude correction coefficient is used as the amplitude-frequency response curve of the filter to be designed, and a finite impulse response digital filter is designed by using a frequency sampling method, so that the amplitude-frequency response curve of the filter is consistent with the obtained amplitude correction coefficient;

[0009] The balance bears the tension of the rotor system. The data acquisition equipment is used to collect the output signal of the balance sensor, which is multiplied by the obtained engineering quantity conversion coefficient to convert it into tension. The tension signal is then filtered with the designed filter. After filtering, an accurate tension measurement value is obtained.

[0010] Preferably, before obtaining a set of reference values ​​of the transfer function by comparing the amplitudes at other frequencies with the DC amplitude based on the obtained transfer function and taking the DC amplitude of the transfer function as a reference, the method further comprises:

[0011] The transfer function is measured on the floating balance in a direction perpendicular to the floating frame, thereby obtaining: when the same amplitude of force is loaded on the balance, the response amplitude of the balance at different frequencies is obtained.

[0012] Preferably, the transfer function is measured on the floating balance in a direction perpendicular to the floating frame to obtain: when the same amplitude of force is loaded on the balance, the response amplitude of the balance at different frequencies, and further includes:

[0013] By adding weights to the floating frame, the engineering quantity conversion coefficient between the balance tension and the sensor output voltage signal when the balance is subjected to static tension is calibrated.

[0014] Preferably, the DC amplitude is a reference "1".

[0015] Preferably, the transfer function is measured in a direction perpendicular to the floating frame while the balance is floating, so as to obtain: when the force of the same amplitude is loaded on the balance, the response amplitude of the balance at different frequencies includes:

[0016] On the floating balance, the transfer function is measured by the hammer method in the direction perpendicular to the floating frame, thereby obtaining: when the same amplitude of force is loaded on the balance, the response amplitude of the balance at different frequencies.

[0017] Preferably, the transfer function is measured in a direction perpendicular to the floating frame while the balance is floating, so as to obtain: when the force of the same amplitude is loaded on the balance, the response amplitude of the balance at different frequencies includes:

[0018] On the floating balance, in the direction perpendicular to the floating frame, the transfer function is measured using the sinusoidal steady-state excitation method, thereby obtaining: when the same amplitude of force is loaded on the balance, the response amplitude of the balance at different frequencies is obtained.

[0019] Preferably, the reference value represents the proportional relationship between the response amplitude of the balance and the DC amplitude when the excitation force amplitude remains unchanged and the frequency changes.

[0020] Preferably, the response amplitude at 10 Hz is three times the DC amplitude, which means that when the excitation amplitude is the same, the response amplitude at 10 Hz is three times the response amplitude at DC.

[0021] This application has the following technical effects:

[0022] For a long time, balances have been calibrated statically by loading weights. This method is accurate when measuring static DC force, but inaccurate when measuring AC dynamic force.

[0023] The invention realizes accurate measurement of dynamic AC force by the balance by measuring the transfer function and designing a filter on the basis of calibration of the balance loaded with weights, thereby making the rotor pulling force measured by the balance during helicopter rotor test more accurate. BRIEF DESCRIPTION OF THE DRAWINGS

[0024] Figure 1 : Schematic diagram of rotor strain balance structure;

[0025] Figure 2 : Schematic diagram of reference values ​​of transfer function;

[0026] Figure 3 : Schematic diagram of amplitude correction coefficient. DETAILED DESCRIPTION

[0027] See also Figure 1-Figure 3 Since the existing balance calibration method uses the method of loading weights, this method can only measure static tension but not dynamic tension. In order to overcome the shortcomings of the existing balance calibration method, the present invention proposes a dynamic calibration method, through which the balance can accurately measure dynamic tension. The specific method is as follows:

[0028] (1) First, according to the existing practice, the balance is statically calibrated by loading weights to obtain the corresponding relationship between the tension and the balance output signal when the balance is subjected to static DC tension. It should be noted that the balance is a complex mechanism with multiple force-bearing rods. It takes into account the situation that the balance is simultaneously subjected to vertical tension, horizontal lateral force, and multi-component coupling of the bending moment around each coordinate axis. Through reasonable design, each force and bending moment is decomposed to reduce the mutual coupling interference. Therefore, the tension finally measured by the balance is calculated by repeated iterations of multiple sensor signals.

[0029] (2) Then, the balance is regarded as a linear time-invariant system, whose input is the tension on the balance, and the output is the tension measurement value calculated based on the calculation formula between the tension obtained when the weight is loaded and calibrated and the sensor output signal. For the linear time-invariant system of the balance, the transfer function is measured by the hammer method or the exciter method. When measuring the transfer function, the input signal is the excitation force, which is loaded vertically to the floating frame of the balance. Through the transfer function, we can get: when the same amplitude of force is loaded on the balance, the response amplitude of the balance at different frequencies.

[0030] (3) For the transfer function obtained in step 2, take its DC amplitude as the reference, and compare the amplitudes at other frequencies with the DC amplitude to obtain a set of reference values ​​of the transfer function. This set of reference values ​​represents the proportional relationship between the response amplitude of the balance relative to the DC excitation at the same excitation force amplitude and different frequencies. If the reference value at a certain frequency is greater than 1, it means that the balance has an amplifying effect on the excitation force at that frequency. If the reference value at a certain frequency is less than 1, it means that the balance has a reducing effect on the excitation force at that frequency. Only when the reference value is 1 can it be said that the balance can accurately measure the force at that frequency.

[0031] (4) Based on the transfer function reference value obtained in step 3, the inverse of the reference value is taken at each frequency to obtain a set of values, called amplitude correction coefficients. When the balance is excited by forces of different frequencies, the balance response amplitude will sometimes be amplified and sometimes reduced according to its transfer function. If the response amplitude of the balance is multiplied by this set of correction coefficients accordingly, the excitation forces of different frequencies can be accurately measured.

[0032] (5) Based on the amplitude correction coefficient obtained in step 4, a finite impulse response digital filter is designed using the frequency sampling method so that the amplitude-frequency response curve of the filter is consistent with the amplitude correction coefficient obtained in step 4. The filter can amplify or reduce the different frequency components in the balance measurement signal by appropriate multiples, so that the balance can accurately measure dynamic force.

[0033] (6) During actual use, the balance is subjected to the tension from the rotor system. The signal of each sensor of the balance is collected by the acquisition device at a set sampling frequency, and the time domain data of the tension of the balance is obtained through the calculation formula. This set of data is then filtered by the filter designed in step 5. After filtering, an accurate tension measurement value is obtained.

[0034] Key Points:

[0035] (1) Measurement of balance transfer function: The balance is regarded as a linear time-invariant system, with the excitation force as input and the calculation formula obtained by static weight calibration. The balance tension measurement value calculated by this formula is output, and its transfer function is measured by hammer method or other methods.

[0036] (2) Filter design: Based on the balance transfer function, the DC amplitude is used as the reference. The amplitudes of other frequencies are compared with the DC amplitude to obtain a set of reference values. The inverse of this set of reference values ​​is then used as the amplitude-frequency response curve of the filter to design the corresponding finite impulse response digital filter.

[0037] For a long time, balances have been calibrated statically by loading weights. This method is accurate when measuring static DC force, but inaccurate when measuring AC dynamic force.

[0038] The invention realizes accurate measurement of dynamic AC force by the balance by measuring the transfer function and designing a filter on the basis of calibration of the balance loaded with weights, thereby making the rotor pulling force measured by the balance during helicopter rotor test more accurate.

[0039] Figure 1 : Schematic diagram of the rotor strain balance structure. The parts numbered 1, 2, 3, and 4 in the figure correspond to the four force-bearing rods of the balance in the vertical direction. There is a force sensor (numbered 5) under each rod. The top of the balance numbered 6 is a floating frame;

[0040] Figure 2 : Schematic diagram of reference values ​​of transfer function, in which the DC amplitude is the reference "1" and the response amplitude at 10Hz is 3 times the DC amplitude;

[0041] Figure 3 : Schematic diagram of amplitude correction coefficient. In the figure, the DC amplitude is the reference "1" and the amplitude correction coefficient at 10Hz is 1 / 3.

[0042] The present application provides a rotor strain balance dynamic performance calibration method, specifically, comprising the following steps:

[0043] Step 1: By adding weights to the floating frame, calibrate the engineering quantity conversion coefficient between the balance tension and the sensor output voltage signal when the balance is subjected to static tension.

[0044] Step 2: On the floating balance, in the direction perpendicular to the floating frame, use the hammer method or the sinusoidal steady-state excitation method to measure the transfer function, so as to obtain: when the same amplitude of force is loaded on the balance, the response amplitude of the balance at different frequencies

[0045] Step 3: Based on the transfer function obtained in step 2, the DC amplitude of the transfer function is used as a reference, and the amplitudes at other frequencies are compared with the DC amplitude to obtain a set of reference values ​​of the transfer function (see Figure 2 ), this set of reference values ​​represents the proportional relationship between the response amplitude of the balance and the DC amplitude when the excitation force amplitude remains unchanged and the frequency changes. Figure 2In the schematic curve of the transfer function reference value given in , the DC amplitude is the reference "1", and the response amplitude at 10 Hz is 3 times the DC amplitude, which means that when the excitation amplitude is the same, the response amplitude at 10 Hz is 3 times the response amplitude at DC.

[0046] Step 4: Based on the transfer function reference value obtained in step 3, take the reciprocal of the reference value at each frequency to obtain a set of values, which are called the amplitude correction coefficients at different frequencies (see Figure 3 ). Taking 10Hz as an example, the reference value of the transfer function at 10Hz is 3, and the reciprocal is 1 / 3. Therefore, the amplitude correction coefficient at 10Hz is 1 / 3, and the same treatment is applied to other frequencies.

[0047] Step 5: Use the amplitude correction coefficient obtained in step 4 as the amplitude-frequency response curve of the filter to be designed, and use the frequency sampling method to design a finite impulse response digital filter so that the amplitude-frequency response curve of the filter is consistent with the amplitude correction coefficient obtained in step 4. Taking 10 Hz as an example, its amplitude correction coefficient is 1 / 3, so the amplitude of the filter is 1 / 3 at 10 Hz, and the filter can attenuate the 10 Hz signal to 1 / 3.

[0048] Step 6. In actual use, the balance is subjected to the tension of the rotor system. The data acquisition equipment is used to collect the output signal of the balance sensor, and this signal is multiplied by the engineering quantity conversion coefficient obtained in step 1 to convert it into tension. The tension signal is then filtered using the filter designed in step 5. After filtering, an accurate tension measurement value is obtained.

Claims

1. A rotor strain balance dynamic performance calibration method, characterized in that: The method comprises the following steps: On the basis of the obtained transfer function, taking the DC amplitude of the transfer function as a reference, the amplitudes at other frequencies are compared with the DC amplitude to obtain a set of reference values ​​of the transfer function; According to the obtained transfer function reference value, the reciprocal of the reference value is taken at each frequency to obtain a set of values, which are called amplitude correction coefficients at different frequencies; The obtained amplitude correction coefficient is used as the amplitude-frequency response curve of the filter to be designed, and a finite impulse response digital filter is designed by using a frequency sampling method, so that the amplitude-frequency response curve of the filter is consistent with the obtained amplitude correction coefficient; The balance bears the tension of the rotor system. The data acquisition equipment is used to collect the output signal of the balance sensor, which is multiplied by the obtained engineering quantity conversion coefficient to convert it into tension. The tension signal is then filtered with the designed filter. After filtering, an accurate tension measurement value is obtained.

2. The method according to claim 1, characterized in that: Before obtaining a set of reference values ​​of the transfer function by comparing the amplitudes at other frequencies with the DC amplitude based on the obtained transfer function and taking the DC amplitude of the transfer function as a reference, the method further includes: The transfer function is measured on the floating balance in a direction perpendicular to the floating frame, thereby obtaining: when the same amplitude of force is loaded on the balance, the response amplitude of the balance at different frequencies is obtained.

3. The method according to claim 2, characterized in that The transfer function is measured on the floating balance in a direction perpendicular to the floating frame, thereby obtaining: when the same amplitude of force is loaded on the balance, the response amplitude of the balance at different frequencies also includes: By adding weights to the floating frame, the engineering quantity conversion coefficient between the balance tension and the sensor output voltage signal when the balance is subjected to static tension is calibrated.

4. The method according to claim 3, characterized in that The DC amplitude is referenced to "1".

5. The method according to claim 4, characterized in that The transfer function is measured on the floating balance in a direction perpendicular to the floating frame, thereby obtaining: when the same amplitude of force is loaded on the balance, the response amplitude of the balance at different frequencies includes: On the floating balance, the transfer function is measured by the hammer method in the direction perpendicular to the floating frame, thereby obtaining: when the same amplitude of force is loaded on the balance, the response amplitude of the balance at different frequencies.

6. The method according to claim 4, characterized in that The transfer function is measured on the floating balance in a direction perpendicular to the floating frame, thereby obtaining: when the same amplitude of force is loaded on the balance, the response amplitude of the balance at different frequencies includes: On the floating balance, in the direction perpendicular to the floating frame, the transfer function is measured using the sinusoidal steady-state excitation method, thereby obtaining: when the same amplitude of force is loaded on the balance, the response amplitude of the balance at different frequencies is obtained.

7. The method according to claim 4, characterized in that The reference value indicates the proportional relationship between the response amplitude of the balance and the DC amplitude when the excitation force amplitude remains unchanged and the frequency changes.

8. The method according to claim 7, characterized in that The response amplitude at 10Hz is three times the DC amplitude, which means that when the excitation amplitude is the same, the response amplitude at 10Hz is three times the response amplitude at DC.

Images