A super-wide-range segment torque calibration method

Abstract

The present application relates to a kind of super-range segment torque calibration methods, for the torque calibration of the maximum range of the torque calibration of the conventional torque measurement standard device beyond the maximum range of the torque calibration of the required calibration torque range section, using linear fitting extrapolation method, first, by separately calculating torque standard value part and torque measurement deviation part, then by different extrapolation coefficients, the calculation results of the two parts are adjusted, i.e. the linear extrapolation calculation value after the extension of the range can be obtained.The super-range segment torque calibration method of the present application can reasonably extrapolate and estimate the torque value of the super-range when calibrating the torque value in the small range.The method can be widely used in the super-torque measurement calibration scene of high-speed engine or wind turbine, to ensure the accurate measurement and traceability of super-torque value in the industrial field.

Description

Technical Field

[0001] This invention relates to the field of torque measurement, and more specifically to a method for calibrating torque over an ultra-large range. Background Technology

[0002] The practical need to measure ultra-high torque values ​​in high-speed engines and wind turbines presents a significant technical challenge: how to accurately measure and calibrate ultra-high torque values ​​that exceed the range of existing torque measurement standards. This issue is particularly prominent in the industrial and energy sectors, as equipment in these environments often requires stable operation under high torque conditions, placing extremely high demands on the accuracy and reliability of ultra-high torque measurement. Summary of the Invention

[0003] The technical problem solved by this invention is to accurately measure and calibrate ultra-large torque values ​​exceeding the range of existing torque measurement standards. To address the limitation of existing torque measurement standards in their design principles and mechanical structures, which prevent direct measurement, calibration, and verification of ultra-large torque values ​​beyond their range, this invention provides an ultra-large range torque calibration method. This method employs a novel linear extrapolation approach, enabling reasonable extrapolation estimation of ultra-large range torque values ​​while calibrating torque values ​​within a small range. This method can be widely applied in ultra-large torque measurement and calibration scenarios such as high-speed engines or wind turbines, ensuring accurate measurement and traceability of ultra-large torque values ​​in industrial applications.

[0004] To achieve the above objectives, the technical solution of the present invention is: an ultra-large range torque calibration method, used for torque calibration where the torque range to be calibrated exceeds the maximum range that can be calibrated by a traditional torque measurement standard device. The method employs a linear fitting extrapolation method. First, the standard torque value and the torque measurement deviation are calculated separately. Then, the calculation results of the two parts are adjusted by different extrapolation coefficients to obtain the linear extrapolation calculation value after the range is expanded.

[0005] Furthermore, the specific calculation method for the linear fitting extrapolation method is as follows:

[0006]

[0007] In equation (1): y i ex —Extrapolation result of the i-th torque value;

[0008] —The standard value corresponding to the i-th torque measurement value;

[0009] m l —The linear extrapolation coefficient between the extrapolation range and the measurement range;

[0010] k l—Linear extrapolation coefficient for the torque measurement deviation portion;

[0011] Δx i —The percentage of error offset for the i-th torque value satisfies the following formula:

[0012]

[0013] In equation (2): x i —The i-th torque measurement value;

[0014] —The standard value corresponding to the i-th torque measurement value;

[0015] As can be seen from equation (1), the key to the linear fitting extrapolation method lies in the extrapolation coefficient m. l and k l The selection of m, where m l The value can be directly determined by the magnification factor between different ranges, but k l The amplitude is determined by the amplitude parameters of a partial segment and the ratio of the full range amplitude to the amplitude of a partial segment; usually, this amplitude ratio is unknown, but is estimated by measuring different partial segments.

[0016] Furthermore, based on the linear fitting extrapolation method, by obtaining multiple amplitude values ​​and linearly extrapolating the amplitude across the entire range, and then using the extrapolation estimation curve as a help, k can be roughly obtained. l The value of m is used to determine the extrapolation coefficient. l and k l Then, the extrapolation estimate can be calculated.

[0017] Furthermore, the torque is measured using an ultra-large torque sensor installed on a torque standard device, resulting in torque measurement curves for different range segments.

[0018] Furthermore, by adjusting the calculation results of the two parts using different extrapolation coefficients, a comparison was obtained of the linear fitting error ratio curves of torque measurement after linear scaling extrapolation for different range segments.

[0019] The beneficial effects of this invention are as follows:

[0020] 1. It breaks through the technical bottleneck that traditional torque measurement standard devices cannot directly calibrate ultra-large torque values ​​that exceed the measurement range;

[0021] 2. The formula is simple and quick to calculate, and easy to operate and implement;

[0022] 3. High calibration accuracy: During the calibration process for the 2.5MN·m ultra-large range torque sensor, the maximum deviation between the estimated value and the standard value did not exceed 0.5% FS. Attached Figure Description

[0023] Figure 1 This is a schematic diagram of mounting a 2.5MN·m ultra-large torque sensor on a 600kN·m torque standard device;

[0024] Figure 2 Torque measurement curves for the 100 kN·m and 500 kN·m ranges;

[0025] Figure 3 A comparison of the percentage curves of linear fitting error in torque measurement after linear scaling and extrapolation for the 100kN·m and 500kN·m ranges. Detailed Implementation

[0026] The present invention will be further described below with reference to the accompanying drawings and embodiments.

[0027] This invention provides a torque calibration method for an ultra-large range. By calibrating the torque sensor within a smaller range and then using linear extrapolation, a reasonable estimate of the torque value for a larger range is made, thereby achieving torque calibration for an ultra-large range. The torque range to be calibrated exceeds the maximum calibrable range of traditional torque measurement standards. Specifically, the torque value to be measured often needs to reach at least 1.5 MN·m, and under certain conditions, it may even reach 5 MN·m or higher, all exceeding the calibration capabilities of current torque standard devices.

[0028] The linear extrapolation method first relies on the calibration capability of traditional torque measurement standards to complete calibration within the sensor's smaller measurement range. Then, by determining the range amplification factor and the linear extrapolation coefficients for different measurement points, linear extrapolation is performed based on a specific formula to calculate the estimated torque value for the ultra-large measurement range.

[0029] Based on the aforementioned technical solution, taking a 2.5MN·m ultra-large torque sensor as an example, the ultra-large range torque calibration method of the present invention is applied. The calibration test process and extrapolation estimation calculation results are as follows:

[0030] First, the 2.5MN·m ultra-large torque sensor was installed on a 600kN·m torque standard device, such as... Figure 1 As shown, torque calibration was carried out in the 100 kN·m and 500 kN·m ranges, with 10 measuring points selected in each range:

[0031] The torque settings for the 0-100 kN·m range are 10 kN·m, 20 kN·m, 30 kN·m, 40 kN·m, 50 kN·m, 60 kN·m, 70 kN·m, 80 kN·m, 90 kN·m and 100 kN·m, respectively.

[0032] The torque settings for the 0-500 kN·m range are 50 kN·m, 100 kN·m, 150 kN·m, 200 kN·m, 250 kN·m, 300 kN·m, 350 kN·m, 400 kN·m, 450 kN·m and 500 kN·m, respectively.

[0033] The full range of 0-2.5MN·m is calculated by extrapolation based on the range calibration data of 0-500kN·m. The three ranges mentioned above are each 5 times each. Therefore, according to formula (1), the coefficients extrapolated from the 100kN·m range to the 500kN·m range can be used to make an approximate extrapolation estimate of the 500kN·m range to the 2.5MN·m range.

[0034] According to the ultra-large range torque calibration method, during the small range calibration of the 2.5MN·m ultra-large torque sensor, each test point was stopped until the instrument reading stabilized. The clockwise and counterclockwise loading and unloading were performed three times consecutively, and the measurement data at each point were recorded. The data was not zeroed during the measurement. In data processing, the three sets of repeated test data were averaged after removing zeros, linearly fitted, and the measurement error ratio was calculated according to formula (2), and curves were plotted. The torque measurement curves of the 2.5MN·m ultra-large torque sensor in the 100kN·m and 500kN·m ranges are shown below. Figure 2 As shown, the torque measurement linear fitting error ratio curve is as follows: Figure 3 As shown:

[0035] Considering that the 600kN·m measuring device has a large error in the 100kN·m range, especially in the ±10-50kN·m measurement range, and that the trends of clockwise and counterclockwise loading in the latter half of the measurement are basically consistent, reflecting the changing trend of measurement error between the large and small ranges, the linear scaling calculation method is used to extrapolate the measurement results in the 500kN·m range, according to formula (1):

[0036]

[0037] Among them, the range is extended from 500 kN·m to 2.5 MN·m. l That is, the linear extrapolation coefficient between the extrapolated range and the measured range is 5, k lThe linear extrapolation coefficient for the torque measurement deviation is calculated based on the conclusion of linear scaling of the curve trend between the 100 kN·m and 500 kN·m ranges. The scaling is performed on the proportion of the fitting error between the measurement and calibration results in the 500 kN·m range. Specifically, the linear extrapolation coefficient for the clockwise portion is 0.2068 (meaning the fitting error proportion after a 5-fold range magnification is 0.2068 times that before magnification, the same applies below), and the linear extrapolation coefficient for the counter-clockwise portion is 0.2145. The specific calculation results are shown in the table below.

[0038] Extrapolation estimation results of linear scaling extrapolation from the 500 kN·m range to the ultra-large range of 2.5 MN·m

[0039]

[0040] Note: Clockwise torque is recorded as a positive number, and counterclockwise torque is recorded as a negative number.

[0041] According to the data in the table, the extrapolation estimation results of each measuring point are very close to the corresponding torque standard values ​​in the table. The maximum deviation between the extrapolation estimation value and the standard value does not exceed 0.5% FS. This shows that the linear extrapolation estimation method works well in the 2.5MN·m ultra-large torque sensor.

Claims

1. A 1.5 MN·m and above range segment torque calibration method, for torque calibration of a required calibration torque range exceeding the maximum range of a 600 kN·m torque metrology standard device, characterized in that: The linear fitting extrapolation method is used, first, the torque standard value part and the torque measurement deviation part are calculated separately, and then the calculation results of the two parts are adjusted by different extrapolation coefficients, so that the linear extrapolation calculation value after the range expansion is obtained; the specific calculation method of the linear fitting extrapolation method is as follows: （1） In formula (1): — the i-th torque value extrapolation result; — the i-th torque measurement value corresponding standard value; — the linear extrapolation coefficient of the extrapolation range and the measurement range; — the linear extrapolation coefficient of the torque measurement deviation part; — the error offset proportion of the i-th torque value, satisfying the following formula: （2） In formula (2): - the i-th torque measurement value; - the standard value corresponding to the i-th torque measurement value, From formula (1), the key of the linear fitting extrapolation method is the selection of the extrapolation coefficient and , wherein, The value can be directly determined by the amplification ratio between different ranges, The value is estimated based on the measurement data of multiple partial sections by the linear fitting extrapolation method. According to the linear fitting extrapolation method, the torque value is first measured in multiple partial sections, the error offset ratio curve is drawn, and then the extrapolation coefficient of the full range section is estimated according to the trend of the curve in the multiple partial sections .

2. The 1.5 MN-m and above range segment torque calibration method of claim 1, wherein: The torque measurement is measured by a 2.5MN·m torque sensor installed on the torque standard device, and the torque measurement curves of different range sections are obtained.

3. The 1.5 MN-m and above range segment torque calibration method of claim 1, wherein: The calculation results of the two parts are adjusted by different extrapolation coefficients, and the comparison of the torque measurement linear fitting error proportion curves of different range sections after linear scaling and extrapolation is obtained.

Images