Pressing plate sensor device and calibration system thereof

Abstract

The invention provides a pressure plate sensor device and a calibration system thereof, and relates to the technical field of pressure plate sensor application, and the calibration system comprises a trend drift fitting module for quantitatively separating trend drift and random drift, a random drift evaluation module, and a trend drift removal module for removing the trend drift. According to the invention, the working time length, the temperature and the humidity are firstly combined and analyzed to quantify trend drift, then the trend drift is removed from an angular velocity measurement value to obtain an angular velocity calibration initial value, and then the angular velocity calibration initial value is processed based on a window averaging method. The angular velocity calibration final value of the smooth random drift is obtained, so that the random error in the angular velocity measurement value is comprehensively removed, the purpose of accurately calibrating the output value of the pressure plate sensor is achieved, and the smooth window with the minimum length is used during the smooth random drift. And the pressing plate sensor can accurately and efficiently identify the opening and closing states of the pressing plate on the power grid.

Description

Technical Field

[0001] This invention relates to the field of pressure plate sensor application technology, specifically a pressure plate sensor device and its calibration system. Background Technology

[0002] As a crucial component of the power grid, the accurate and effective monitoring of the opening and closing status of pressure plates is of great significance. For example, by monitoring the status of pressure plates, abnormalities can be detected promptly, preventing safety accidents caused by misoperation or equipment failure. It can also quickly locate fault areas, reduce power outage time, and improve fault response speed. Currently, pressure plate sensors equipped with gyroscopes are commonly used to monitor the opening and closing status of pressure plates. However, gyroscopes are susceptible to external influences and therefore have a certain degree of error. Therefore, effective calibration of the gyroscope is particularly important for pressure plate monitoring.

[0003] For gyroscopes, the errors mainly include deterministic errors and random errors. Deterministic errors are systematic errors that can be eliminated through calibration and other processes. Random errors, however, have no obvious pattern and are the main bottleneck limiting the improvement of gyroscope accuracy. In the existing technology, a real-time self-calibration system and method for gyroscopes (main classification number G01C) with authorization announcement number "CN114689082B" calculates the output of the gyroscope after calibration by coefficients K and B at certain time intervals when the gyroscope is in a stationary state (zero input). An array of a certain length is set to store the output of the gyroscope after calibration. The values ​​of this array can be filled sequentially or updated cyclically. The mean value of the array data is used as an estimate of the random error of the gyroscope and subtracted in real time to obtain a high-precision attitude angle, thereby achieving the purpose of eliminating random errors.

[0004] However, existing technologies still have significant shortcomings. For example, random errors are mainly composed of trend drift and random drift. Trend drift refers to the gradually changing bias of the sensor output signal over a certain period of time, which has a certain time dependence. Random drift, on the other hand, refers to the irregular offset of the sensor output signal over a certain period of time, which does not have time dependence. Existing technologies directly use the window averaging method to smooth random errors without taking into account the time-dependent trend drift, resulting in unsatisfactory calibration results.

[0005] The information disclosed in the background section is only intended to enhance the understanding of the background of this disclosure, and therefore may include information that does not constitute prior art known to those skilled in the art. Summary of the Invention

[0006] The purpose of this invention is to provide a pressure plate sensor device and its calibration system to solve the problems mentioned in the background art.

[0007] To achieve the above objectives, the present invention provides the following technical solution: A calibration system for calibrating a gyroscope in a pressure plate sensor device, comprising: Gyroscope testing equipment is used to test gyroscopes to obtain angular velocity measurements of the gyroscopes under various test scenarios, including the actual angular velocity value, operating temperature, and operating humidity. The trend drift fitting module is used to analyze the measured and true angular velocity values ​​under the test scenario to determine the random error value of the angular velocity. It also incorporates the working temperature and humidity of the test scenario and solves the functional expression of the trend drift fitting value based on the gradient descent method. The random drift evaluation module is used to combine the functional expression of the trend drift fitting value and the random error value of the gyroscope in the changing scenario to construct a scatter plot of working time-random drift value, and optimize the functional expression of the trend drift fitting value based on the scatter plot until the scatter plot meets the randomness standard. The trend drift removal module generates a trend drift fitting value of the gyroscope to be calibrated in the pressure plate sensor device under the current sampling based on the optimized function expression, and subtracts the trend drift fitting value from the angular velocity measurement value under the current sampling to obtain the initial value of angular velocity calibration. The random drift smoothing module is used to analyze the trend drift fitting value, operating temperature and operating humidity of the gyroscope to be calibrated under the current sampling to determine the minimum length of the smoothing window. Based on this minimum length, the initial value of angular velocity calibration is processed by the window averaging method to obtain the final value of angular velocity calibration.

[0008] Furthermore, the gyroscope testing equipment consists of a precision rotating platform, a temperature and humidity control device, and data acquisition software, wherein: A precision rotary platform is used to provide stable rotational motion for detecting gyroscopes at different angular velocities; Temperature and humidity control device is used to provide stable temperature and humidity during the measurement process of gyroscope, so as to verify the working performance of gyroscope under different operating temperatures / humidities; Data acquisition software is used to record angular velocity measurements, actual angular velocity values, operating temperature, and operating humidity during the integrated testing process.

[0009] Furthermore, the functional expression for the trend drift fitted value is as follows:

[0010]

[0011] In the formula, Indicates the first The angular velocity measurement value obtained during the second sampling. Indicates the first The true value of the angular velocity corresponding to the next sample. Indicates the gyroscope at the 1st Random error values ​​generated during each sampling. Index of the number of samples; In the formula, Indicates the gyroscope at the 1st Trend drift fitted value at the second sampling time These represent the gyroscope at the [number]th [time]. The duration of each sampling affects the fitted value, as do temperature and humidity.

[0012] Furthermore, the method for fitting and solving the functional expression of the trend drift fitted value is as follows: 1) Set up a standard test scenario, select an unused gyroscope and test it in the standard test scenario to obtain the angular velocity measurement value of the gyroscope in the standard test scenario; 2) Calculate the random error value of the gyroscope at each sampling time in the standard test scenario, and record the working time of the gyroscope at each sampling time. Construct a scatter plot of working time-random error value in the standard test scenario. Analyze the scatter plot based on data processing software. Use the determination coefficient being in the preset range of the determination coefficient as a constraint condition, and solve the functional expression of the influence of time on the fitted value by gradient descent method. 3) Obtain the angular velocity measurement value of the gyroscope under the standard humidity test scenario, and obtain the corresponding random error value and working time. Construct a scatter plot of working time-random error value under the standard humidity test scenario. Analyze the scatter plot based on data processing software. Use the determination coefficient being within the preset range of the determination coefficient as a constraint condition. Solve the functional expression of the temperature influence fitting value through the gradient descent method. The standard humidity test scenario is a test scenario with the same working humidity as the standard test scenario. 4) Obtain the angular velocity measurement value of the gyroscope in the test scenario, and obtain the corresponding random error value and working time. Construct a scatter plot of working time-random error value in the test scenario. Analyze the scatter plot based on data processing software. Use the determination coefficient being in the preset range of the determination coefficient as a constraint condition, and solve the functional expression of the humidity effect on the fitted value by gradient descent method.

[0013] Furthermore, the random drift value is the difference between the random error value and the trend drift fitted value; The condition for a scatter plot to satisfy the randomness criterion is that, within any working time period, the autocorrelation coefficient of the random drift value is within a preset zero-reaching interval for all lag periods. The method for optimizing the function expression of the trend drift fitting value is as follows: the function expression of the trend drift fitting value output by the trend drift fitting module is used as the initial function expression. When the scatter plot does not meet the randomness standard, the gradient descent method is used to update the initial function expression until the scatter plot meets the randomness standard. When optimizing the function expression, it is ensured that the optimized function expression meets the constraint that the coefficient of determination is within the preset range of the coefficient of determination.

[0014] Furthermore, the minimum length of the smoothing window is obtained based on the smoothing accuracy analysis model. The smoothing accuracy analysis model built on the TensorFlow deep learning framework includes an input layer for receiving trend drift fitting values, working temperature, working humidity and smoothing window length, multiple hidden layers for processing the input data, with ReLU selected as the activation function on the hidden layers, and an output layer for outputting the smoothing relative error. The calculation logic for the smoothing relative error is as follows: the window averaging method is used to obtain the final value of the angular velocity calibration, then the ratio of the final value of the angular velocity calibration to the true value of the angular velocity is calculated as the smoothing accuracy, and finally the absolute difference between the smoothing accuracy and 1 is calculated as the smoothing relative error.

[0015] Furthermore, the logic for obtaining the minimum length of the smoothing window under the current sampling is as follows: input the trend drift fitting value, working temperature, working humidity, and the minimum length of the smoothing window under the previous sampling into the smoothing accuracy analysis model after training, and keep the trend drift fitting value, working temperature, and working humidity unchanged. Use the smoothing relative deviation output by the smoothing accuracy analysis model being less than a preset threshold as a constraint condition, and gradually adjust the minimum length of the smoothing window under the previous sampling until the minimum value is found as the minimum length of the smoothing window of the gyroscope to be calibrated under the current sampling.

[0016] A pressure plate sensor device for detecting the open / closed state of a pressure plate on a power grid includes a gyroscope to be calibrated using the aforementioned calibration system, a pressure plate sensor body, and a clamping mechanism, wherein: The pressure plate sensor body includes a pressure plate sensor housing, a circuit board fixed inside the pressure plate sensor housing, an antenna, and a battery. The circuit board is used to acquire the opening and closing state of the pressure plate on the power grid, and the gyroscope to be calibrated is mounted on the circuit board. The antenna is used to transmit the opening and closing state of the pressure plate on the power grid to the outside world, and the battery is used to power the circuit board. A clamping mechanism is fixedly connected to the surface of the pressure plate sensor housing, and the pressure plate sensor housing and the pressure plate are detachably connected through the clamping mechanism.

[0017] Furthermore, the clamping mechanism consists of at least two thin plates fixedly connected to the surface of the pressure plate sensor housing. By inserting the thin plates into the slots of the pressure plate, a detachable connection between the pressure plate sensor housing and the pressure plate is achieved.

[0018] Furthermore, the clamping mechanism is a recessed bracket with an opening that is fixedly connected to the surface of the pressure plate sensor housing. By inserting the pressure plate into the recessed bracket, a detachable connection between the pressure plate sensor housing and the pressure plate is achieved.

[0019] Compared with the prior art, the beneficial effects of the present invention are: The pressure plate sensor device and its calibration system of the present invention combine analysis of working time and temperature and humidity to quantify trend drift, then remove the trend drift from the angular velocity measurement value to obtain the initial value of angular velocity calibration, and then process the initial value of angular velocity calibration based on the window averaging method to obtain the final value of angular velocity calibration with smooth random drift. In this way, the random error in the angular velocity measurement value is completely removed, and the purpose of accurate calibration of the output value of the pressure plate sensor is achieved. In smoothing random drift, the minimum length of the smoothing window is used, which effectively reduces the delay of online calibration and ensures that the pressure plate sensor accurately and efficiently identifies the opening and closing status of the pressure plate on the power grid. Attached Figure Description

[0020] Figure 1 This is a block diagram of the calibration system in this invention; Figure 2 This is a three-dimensional view of the pressure plate sensor device with a thin sheet in this invention; Figure 3 This is a two-dimensional diagram of the pressure plate sensor device with a thin sheet in this invention; Figure 4 This is a three-dimensional view of the pressure plate sensor device with a concave card seat in this invention; Figure 5 This is a distribution diagram of random error values; Figure 6 This is a distribution diagram of random offset values; Figure 7 This is a distribution diagram showing the accuracy of monitoring the opening and closing status of the pressure plate.

[0021] Figure description: 1-Pressure plate sensor body, 2-Clamping mechanism. Detailed Implementation

[0022] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to specific embodiments.

[0023] It should be noted that, unless otherwise defined, the technical or scientific terms used in this invention should have the ordinary meaning understood by one of ordinary skill in the art to which this invention pertains. The terms "first," "second," and similar terms used in this invention do not indicate any order, quantity, or importance, but are merely used to distinguish different components. Terms such as "comprising" or "including" mean that the element or object preceding the word encompasses the elements or objects listed following the word and their equivalents, without excluding other elements or objects. Terms such as "connected" or "linked" are not limited to physical or mechanical connections, but can include electrical connections, whether direct or indirect. Terms such as "upper," "lower," "left," and "right" are used only to indicate relative positional relationships; when the absolute position of the described object changes, the relative positional relationship may also change accordingly.

[0024] Example 1: Please see Figure 1 , Figure 5-7 This embodiment provides a calibration system for calibrating the gyroscope in a pressure plate sensor device, including: Gyroscope testing equipment is used to test gyroscopes to obtain angular velocity measurements under various test scenarios, including actual angular velocity values, operating temperature, and operating humidity. The equipment consists of a precision rotating platform, temperature and humidity control devices, and data acquisition software. A precision rotating platform is used to provide stable rotational motion for testing gyroscopes at different angular velocities. Here, the angular velocity measured by the gyroscope on the precision rotating platform is defined as the measured angular velocity value, and the angular velocity set by the precision rotating platform at this time is defined as the true angular velocity value. Clockwise is defined as a positive value and counterclockwise as a negative value. By comparing the difference between the measured angular velocity value and the true angular velocity value, the accuracy of the gyroscope output value can be determined, providing a theoretical basis for the calibration operation of the gyroscope in the following text. The precision rotating platform can be a high-precision rotating platform such as the Aerotech AGS100, which is widely used in scenarios for dynamic and static testing of gyroscopes. A temperature and humidity control device is used to provide stable temperature and humidity during the measurement process of the gyroscope to verify the working performance of the gyroscope under different operating temperatures / humidities. This is so that when calibrating the gyroscope later, the influence of temperature and humidity on the working performance of the gyroscope can be fully considered. The temperature and humidity control device can be a constant temperature and humidity chamber of model Binder KB 115 or a high-precision temperature and humidity chamber of model Memmert HPP110. The operating temperature and humidity can be obtained directly from the temperature and humidity setting interface of the above-mentioned temperature and humidity control device. It is assumed here that the output of the precision rotating platform and the temperature and humidity control device is accurate and error-free. Data acquisition software is used to record and integrate the measured angular velocity values, the actual angular velocity values, the operating temperature and the operating humidity at each test moment, so as to perform subsequent data analysis and calibration operations. The data acquisition software can use conventional software such as LabVIEW or MATLAB. It should be noted that when using gyroscope testing equipment, the same model of gyroscope as the one to be calibrated in the pressure plate sensor device should be selected for testing. This is so that the trend drift fitting value function expression for fitting optimization can be directly used for the gyroscope to be calibrated. Moreover, since the gyroscope determines the tilt angle by integrating the angular velocity, and thus determines the opening and closing state of the pressure plate, the angular velocity is calibrated here, rather than the tilt angle.

[0025] The process of testing the gyroscope is as follows: 1) Set multiple true values ​​of angular velocity, operating temperature, and operating humidity, and arrange and combine them to construct multiple test scenarios; As one implementation method, the true angular velocity value can be set to six groups: 0 degrees per second, 30 degrees per second, 60 degrees per second, 90 degrees per second, 120 degrees per second, 150 degrees per second, etc. The operating temperature can be set to five groups: -25 degrees Celsius, 0 degrees Celsius, 25 degrees Celsius, 50 degrees Celsius, and 75 degrees Celsius. Alternatively, operators can set these according to the actual temperature of the pressure plate sensor's operating environment, ensuring coverage of the sensor's normal operating temperature range. The operating humidity can be set to five groups: 0%, 20%, 40%, 60%, and 80%. Again, operators can set these according to the humidity of the pressure plate sensor's actual operating environment, ensuring coverage of the sensor's normal operating humidity range. For example, one test scenario could be set as: true angular velocity value 0 degrees per second, operating temperature 25 degrees Celsius, and operating humidity 20%. 2) Place the gyroscope in each test scenario for performance testing, and collect angular velocity measurements during the test. The specific test process is as follows: The working temperature and humidity in the test scenario are input into the temperature and humidity control device. After the environment created by the temperature and humidity control device meets the standard and stabilizes, the precision rotating platform with the gyroscope is placed in the temperature and humidity control device. The angular velocity output value of the precision rotating platform is set to the true angular velocity value in the test scenario. The precision rotating platform is then started. After the angular velocity output value of the precision rotating platform stabilizes, the angular velocity measurement value output by the gyroscope is collected. It should be noted that when testing the gyroscope for each test scenario, the test duration should not be less than 1 hour, specifically set to 1 hour, 2 hours, 3 hours, etc., to facilitate subsequent analysis of the continuous deviation (i.e., trend drift) of the gyroscope output value over a long period. To improve testing efficiency, multiple gyroscopes can be used for testing simultaneously. For example, test scenarios with the same true angular velocity value can be grouped into the same test scenario combination, resulting in six test scenario combinations. For each test scenario combination, one gyroscope can be used for testing. By setting up six gyroscopes to test simultaneously, the test duration can be shortened to one-sixth of the original, thereby improving testing efficiency.

[0026] The trend drift fitting module is used to analyze the measured and true angular velocity values ​​under the test scenario to determine the random error value of the angular velocity. It also incorporates the working temperature and humidity of the test scenario and solves the functional expression of the trend drift fitting value based on the gradient descent method. The functional expression for the trend drift fitted value is as follows:

[0027]

[0028] In the formula, Indicates the first The angular velocity measurement value obtained during the sampling, i.e., the gyroscope's measurement value at the [number]th sampling point. The output value at the next sampling time Indicates the first The true value of the angular velocity corresponding to the sampling time, i.e., the th sampling time. The output value of the precision rotating platform during the second sampling. Indicates the gyroscope at the 1st The random error value generated during each sampling is composed of both random drift and trend drift, and The larger the value, the higher the gyroscope's value in the [number of seconds]. The larger the random error value generated in the first sampling, the greater the error in the second sampling. The less accurate the subsequent sampling, the... This is the index for the number of samples, and the sampling frequency is set to be consistent with the sampling frequency of the gyroscope to be calibrated; In the formula, Indicates the gyroscope at the 1st Trend drift fitted value at the second sampling time These represent the gyroscope at the [number]th [time]. The duration of sampling affects the fitted value, temperature affects the fitted value, and humidity affects the fitted value. The sum of these three factors gives the gyroscope's fitted value at the first sampling point. Trend drift fitted value at the second sampling time ; It should be noted that trend drift refers to the gradual change in bias of the sensor output signal over a certain period of time. The trend drift of a gyroscope is mainly determined by the changes in the physical properties of the internal materials of the gyroscope due to the current working environment (i.e., current influence) and the mechanical fatigue and material aging of the gyroscope due to past working time and working environment (i.e., historical influence). Existing technologies generally analyze the current influence and historical influence independently. However, when analyzing historical influence, it is difficult to integrate the three factors that jointly affect mechanical fatigue and material aging: working time, working temperature, and working humidity. This results in problems such as difficulty in quantification, complex modeling, and high computational cost. This solution creatively decomposes the trend drift fitted value into three components: the fitted value affected by time, the fitted value affected by temperature, and the fitted value affected by humidity. Then, it performs fitting and solving on the fitted values ​​affected by time, temperature, and humidity in a hierarchical manner to improve the accuracy and efficiency of obtaining the trend drift fitted value. This effectively solves the problems of difficulty in quantification, complex modeling, and high computational cost when solving the trend drift fitted value. The functional expression for how duration affects the fitted value is as follows:

[0029] In this formula, polynomials have good local fitting ability and can handle nonlinear relationships well. Therefore, polynomials are used to fit and solve the time fitting values. Indicates the gyroscope cutoff point. The working time during the sampling, where n is the degree variable in the polynomial, and N represents the total degree of the polynomial, and its specific value is determined by the staff based on the actual situation. Generally, it is between 5 and 8 to avoid overfitting while ensuring accurate fitting. During the fitting process, N can be adjusted in real time based on the coefficient of determination, an evaluation index. For example, N can be increased when the coefficient of determination is small to improve fitting accuracy, and N can be decreased when the coefficient of determination is large to avoid overfitting. express to the power of n, The coefficients are polynomial coefficients, representing In calculation The weight of time Specifically, the solution is obtained by fitting using the gradient descent method, and the coefficients of the same type of polynomial starting from n=1 satisfy the constraint that the cumulative value is equal to 1. The functional expression for the effect of temperature on the fitted values ​​is as follows:

[0030] In the formula, Indicates up to the number During the second sampling, the duration for which the gyroscope's operating temperature deviation value falls within the k-th temperature deviation interval is recorded. The operating temperature deviation value is the difference between the operating temperature and the standard operating temperature. Here, the operating temperature under the standard test scenario is taken. As the standard operating temperature, k is the index of the temperature deviation range, and K represents the total number of temperature deviation intervals. The upper and lower limits of each temperature deviation interval are determined by preset the length of the temperature deviation interval. This represents the median value of the k-th temperature deviation interval, specifically the average of the upper and lower limits of the k-th temperature deviation interval. For the first During each sampling, the gyroscope's operating temperature is set to 0 when it reaches 0 to the power of 0, and to 1 when it reaches 0 divided by 0, in order to avoid the above formula becoming meaningless. It should be noted that the smaller the length of the temperature deviation interval, the more precise the interval division, and the more detailed the representation of the historical impact caused by different temperatures. However, the amount of calculation will also increase significantly. Therefore, the specific value of the temperature deviation interval is set by the staff according to the actual situation. For example, the length of the temperature deviation interval can be 2 degrees Celsius, 5 degrees Celsius, 10 degrees Celsius, etc. For example, when the length of the temperature deviation interval is 10 degrees Celsius, it means that the difference between the upper and lower limits of the temperature deviation interval is 10 degrees Celsius. It should be noted that the temperature deviation intervals determined based on the total number and length of the temperature deviation intervals satisfy the following constraints: 1) The temperature deviation interval includes one special temperature deviation interval and other normal temperature deviation intervals. The upper and lower limits of the special temperature deviation interval are both 0, and the normal temperature deviation intervals are arranged sequentially on both sides of 0, and the normal temperature deviation intervals do not include 0; 2) Up to the [number missing]th temperature deviation interval... During each sampling, any operating temperature deviation value can fall within a corresponding temperature deviation range; for example, setting... There are a total of 12 temperature deviation ranges; In the formula, the fitted value of temperature influence consists of the fitted value of historical temperature influence and the fitted value of current temperature influence. Here, we continue to use a polynomial to fit and solve the fitted value of historical temperature influence and the fitted value of current temperature influence. It should be noted that, Used to characterize up to the [number]th The historical temperature fitting value generated by the working temperature corresponding to the k-th temperature deviation interval during the second sampling is used here to simplify the calculation. To replace all operating temperatures corresponding to the k-th temperature deviation interval in the calculation, in order to avoid... The problem of even powers being positive when the value is negative leads to a logical error in the formula. This issue is addressed here. To ensure the accuracy of the calculation of the fitted values ​​based on the influence of historical temperature, and through... Multiply The settings couple operating temperature and operating time to characterize the contribution of a certain temperature to the gyroscope's trend drift over a certain period of time, and then... Up to the date During each sampling, the historical temperature fitting values ​​generated by the operating temperatures corresponding to all temperature deviation intervals are accumulated to characterize the influence of historical temperatures on the fitting values. The coefficients are polynomial coefficients, representing The weight assigned to the historical temperature fitting value corresponding to the k-th temperature deviation interval. Specifically, the solution is obtained by fitting using the gradient descent method; It should be noted that, Used to characterize the effect of the current temperature on the fitted value, here... to replace The calculations are performed to lay the foundation for the subsequent solution based on the gradient descent method. Provide a foundation, The settings are also for circumvention. , The coefficients are polynomial coefficients, representing The weight given to the current temperature in calculating the fitted value.

[0031] The functional expression for the effect of humidity on the fitted values ​​is as follows:

[0032] In the formula, Indicates up to the number During the second sampling, the gyroscope's operating humidity deviation value was at the [missing value]. The working time is defined as the duration of each humidity deviation range. The working humidity deviation value is the difference between the working humidity and the standard working humidity. Here, the working humidity under the standard test scenario is taken. As standard operating humidity This is the index for the humidity deviation range, and L represents the total number of humidity deviation intervals. The upper and lower limits of each humidity deviation interval are determined by preset the length of the humidity deviation interval. Indicates the first The median value of the humidity deviation range, specifically the first... The average of the upper and lower limits of each humidity deviation range. For the first During the next sampling, the gyroscope's operating humidity is also set to 0 when it is 0 to the power of 0, and to 1 when it is 0 divided by 0, in order to avoid the above formula becoming meaningless. It should be noted that the smaller the length of the humidity deviation interval, the more precise the interval division, and the more accurately it can characterize the historical impact of humidity. However, the amount of calculation will also increase significantly. Therefore, the specific length of the humidity deviation interval is set by the staff according to the actual situation. For example, the length of the humidity deviation interval can be 2%, 5%, 10%, etc. For instance, when the length of the humidity deviation interval is 10%, it means that the difference between the upper and lower limits of the humidity deviation interval is 10%. It should be noted that the humidity deviation intervals determined based on the total number and length of the humidity deviation intervals satisfy the following constraints: 1) The humidity deviation interval includes one special humidity deviation interval and other normal humidity deviation intervals. The upper and lower limits of the special humidity deviation interval are both 0. The normal humidity deviation intervals are arranged sequentially on both sides of 0, and the normal humidity deviation intervals do not include 0; 2) Up to the [number missing]th [number missing]... During each sampling, any operating humidity deviation value can fall within a corresponding humidity deviation range; for example, setting... There are a total of 10 humidity deviation ranges; In the formula, the humidity effect fitting value is composed of the historical humidity effect fitting value and the current humidity effect fitting value. Here, we continue to use a polynomial to fit and solve the historical humidity effect fitting value and the current humidity effect fitting value. It should be noted that, Used to characterize up to the [number]th During the second sampling, the first The historical humidity fitting values ​​generated for the working humidity corresponding to each humidity deviation interval are used here to simplify the calculation. to replace the first The calculation is performed for all operating humidity levels corresponding to each humidity deviation range to avoid... The problem of even powers being positive when the value is negative leads to a logical error in the formula. This issue is addressed here. To ensure the accuracy of the calculation of the fitted values ​​for the influence of historical humidity, and through... Multiply The settings couple operating humidity and operating time to characterize the contribution of a certain humidity level to the gyroscope's trend drift over a given time period, and then... Up to the date During each sampling, the historical humidity fitting values ​​generated by the working humidity corresponding to all humidity deviation intervals are accumulated to characterize the influence of historical humidity on the fitting values. The coefficients are polynomial coefficients, representing In calculating the first The weight of the historical humidity fitting value corresponding to each humidity deviation interval. Specifically, the solution is obtained by fitting using the gradient descent method; It should be noted that, Used to characterize the effect of current humidity on the fitted value, here... to replace The calculations are performed to lay the foundation for the subsequent solution based on the gradient descent method. Provide a foundation, The settings are also for circumvention. The problem arises because even powers of negative values ​​are positive, leading to a logical error in the formula. The coefficients are polynomial coefficients, representing The weight given to the current humidity level when calculating the fitted value.

[0033] The method for fitting and solving the above function expressions is as follows: 1) Set up a standard test scenario, select an unused gyroscope and test it in the standard test scenario to obtain the angular velocity measurement value of the gyroscope in the standard test scenario; It should be noted that the standard test scenario preferentially selects a test scenario with a true angular velocity of 0 degrees per second, an operating temperature of 25 degrees Celsius, and an operating humidity of 60%, for the following reasons: 1) In a test scenario with a true angular velocity of 0 degrees per second, the measured angular velocity value is the random error value, eliminating the need to subtract the true angular velocity value and simplifying calculations; 2) An operating temperature of 25 degrees Celsius and an operating humidity of 60% are suitable for the operation of the gyroscope and will not have an excessive impact on it. The trend drift of the gyroscope is mainly affected by the operating time, facilitating subsequent analysis. The fitting solution is then performed; of course, the standard test scenario can also be randomly selected from the above-mentioned multiple test scenarios, which will be determined by the staff according to the actual situation, and will not be elaborated here. 2) Calculate the random error value of the gyroscope at each sampling time in the standard test scenario, and record the working time of the gyroscope at each sampling time. Construct a scatter plot of working time-random error value in the standard test scenario (the working time is used as the independent variable (i.e., the horizontal axis) and the random error value is used as the dependent variable (i.e., the vertical axis) in the scatter plot). Analyze the scatter plot using data processing software (such as the polyfit function in Python or the polyfit function in MATLAB). Use the determination coefficient being within the preset range of the determination coefficient as a constraint condition, and solve the functional expression of the effect of the time on the fitted value using the gradient descent method. It should be noted that under standard testing conditions, the operating temperature remains stable. The gyroscope trend drift component caused by the historical influence of the operating temperature can be approximated as a variable value related to the standard operating temperature and increasing linearly with time. Conversely, the gyroscope trend drift component caused by the current influence of the operating temperature can be approximated as a fixed value related to the standard operating temperature. Since the standard operating temperature is a constant, under standard testing conditions, the gyroscope trend drift component caused by the historical influence of the operating temperature can be approximated as a variable value increasing linearly with time, while the gyroscope trend drift component caused by the current influence of the operating temperature can be approximated as a fixed value. To simplify calculations, when solving for the fitted value of the historical temperature influence, the operating temperature deviation value is first calculated to characterize the operating temperature. Then, multiple temperature deviation intervals are divided, and the median value of each interval is used to uniformly characterize all operating temperatures corresponding to that interval. Similarly, to simplify calculations, when solving for the fitted value of the current temperature influence, a similar approach is used. to replace This setup is intended to account for the historical impact of standard operating temperatures under standard testing conditions. China and Israel as One component, and transfers the current impact caused by the standard operating temperature to China and Israel as One of the components, which is finally solved to obtain and That's it, no need to go from and The temperature-affected component is extracted from the data, i.e., the fitted value of the temperature effect. The function expression only includes a portion of the fitting value affected by the actual temperature, which greatly simplifies the fitting analysis. Similarly, in a standard testing scenario, the operating humidity remains stable. The gyroscope trend drift component caused by the historical influence of the operating humidity can be approximated as a variable value related to the standard operating humidity and increasing linearly with time. The gyroscope trend drift component caused by the current influence of the operating humidity can be approximated as a fixed value related to the standard operating humidity. Since the standard operating humidity is a constant, in a standard testing scenario, the gyroscope trend drift component caused by the historical influence of the operating humidity can be approximated as a variable value increasing linearly with time, while the gyroscope trend drift component caused by the current influence of the operating humidity can be approximated as a fixed value. To simplify the calculation, when solving for the historical humidity influence fitting value, the operating humidity deviation value is first calculated to represent the operating humidity, and then multiple humidity deviation intervals are divided. The median value of each humidity deviation interval is used to uniformly represent all operating humidity values ​​corresponding to that interval. Similarly, to simplify the calculation, when solving for the current humidity influence fitting value, a similar approach is used. to replace This setup is intended to account for the historical effects of standard operating humidity under standard testing conditions. China and Israel as One component, and transfers the current impact caused by standard operating humidity to China and Israel as One of the components, which is finally solved to obtain and That's it, no need to go from and The humidity-affected component is extracted from the data, i.e., the fitted value of the humidity effect. The function expression only includes a portion of the actual humidity affecting the fitted value, which greatly simplifies the fitting analysis. Therefore, under standard testing scenarios, the effects of operating temperature and humidity are quantified as part of the duration-affected fitted value. When constructing the trend drift fitted curve in the scatter plot of operating duration versus random error value, it is only necessary to apply the functional expression of the duration-affected fitted value for fitting and solving. That is, the functional expression of the trend drift fitted curve is: The polynomial coefficients can be solved simply by using the gradient descent method. without needing to solve simultaneously , , and This greatly reduces the difficulty of fitting. Solving for polynomial coefficients using the gradient descent method is common knowledge in the field of science and will not be elaborated here. It should be noted that the random error value is obtained by superimposing the trend drift value and the random drift value. When fitting a trend drift curve on a scatter plot, the trend drift value cannot completely coincide with the random error value due to the influence of the random drift value. That is, the trend drift value on the fitted curve will inevitably have a certain error with the random error value. Therefore, a higher coefficient of determination is not always better when fitting. An excessively high coefficient of determination indicates an overfitting distortion problem. Therefore, it is necessary to limit the coefficient of determination to a preset range to ensure the rationality of the trend drift fitted curve. The specific value of the preset range of the coefficient of determination is set by the staff according to the actual situation, and is generally between 0.75 and 0.85. When calculating the coefficient of determination, the trend drift fitted value is used as the predicted value, and the random error value is used as the true value to calculate the coefficient of determination. The specific calculation process is existing technology and can be obtained through the built-in program of the data processing software, which will not be elaborated here. 3) Obtain the angular velocity measurement value of the gyroscope under the standard humidity test scenario, and obtain the corresponding random error value and working time. Construct a scatter plot of working time-random error value under the standard humidity test scenario. Analyze the scatter plot based on data processing software. Use the determination coefficient being within the preset range of the determination coefficient as a constraint condition, and solve the functional expression of the temperature influence fitting value by gradient descent method. The standard humidity test scenario is a test scenario where the operating humidity is consistent with the standard test scenario, so that the historical impact caused by the standard operating humidity can be accounted for under the standard humidity test scenario. China and Israel as One component, and transfers the current impact caused by standard operating humidity to China and Israel as One of the components; Therefore, in a standard humidity testing scenario, the impact of working humidity is quantified as a part of the temperature influence fitted value. When constructing the trend drift fitting curve in the scatter plot of working time versus random error value, it is only necessary to apply the functional expressions of the working time influence fitted value and the temperature influence fitted value for fitting and solving. That is, the functional expression of the trend drift fitting curve at this time is: And polynomial coefficients As already obtained above, we only need to solve for the polynomial coefficients using the gradient descent method. , without needing to solve simultaneously and This greatly reduces the difficulty of fitting; It should be noted that in step 3), obtaining the random error value and working time, constructing the scatter plot, and solving the function expression of the temperature influence fitting value through data processing software can all be done using the same method as in step 2), which will not be elaborated here. The constraints in step 3) are also the same as those in step 2). 4) Obtain the angular velocity measurement value of the gyroscope in the test scenario, and obtain the corresponding random error value and working time. Construct a scatter plot of working time-random error value in the test scenario. Analyze the scatter plot based on data processing software. Use the determination coefficient being in the preset range of the determination coefficient as a constraint condition, and solve the functional expression of the humidity effect on the fitted value by gradient descent method. In step 4) above, the functional expression of the trend drift fitting curve is: And polynomial coefficients , As already obtained above, we only need to solve for the polynomial coefficients using the gradient descent method. and This greatly reduces the difficulty of fitting; It should be noted that in step 4), obtaining the random error value and working time, constructing the scatter plot, and solving the function expression of the temperature influence fitting value using data processing software can all be done using the same methods as in step 4), and will not be repeated here. Furthermore, the constraints in step 4) are the same as those in step 2). By setting up steps 1) to 4) above, the fitting values ​​of duration, temperature and humidity are fitted in a hierarchical manner to improve the accuracy and efficiency of obtaining trend drift fitting values. This effectively solves the problems of difficulty in quantification, complex modeling and high computational cost when solving trend drift fitting values.

[0034] The random drift evaluation module is used to combine the functional expression of the trend drift fitting value and the random error value of the gyroscope in the changing scenario to construct a scatter plot of working time-random drift value, and optimize the functional expression of the trend drift fitting value based on the scatter plot until the scatter plot meets the randomness standard. It should be noted that the changing scenario is a test scenario in which the actual value of angular velocity, working temperature and working humidity change over time. The gyroscopes tested in both the test scenario and the changing scenario are gyroscopes of the same model as the gyroscope to be calibrated. Among them, the random drift value is the difference between the random error value and the trend drift fitted value. The trend drift fitted value is obtained by analyzing the working time, working temperature and working humidity under changing scenarios based on the functional expression of the trend drift fitted value. The scatter plot satisfies the randomness standard as follows: during any working period, the autocorrelation coefficient of the random drift value under all lag periods is within a preset zero-reaching interval. The zero-reaching interval is an interval that approaches zero. When the autocorrelation coefficient of the random drift value under all lag periods is within the zero-reaching interval, it indicates that the autocorrelation coefficient of the random drift value under all lag periods is close to 0. This further indicates that the time dependence of the random drift value is smaller during that working period, and the more it satisfies the characteristics of the random drift value. This also indicates that the function expression used to calculate the trend drift value is more accurate. The specific value of the zero-reaching interval is set by the staff according to the actual situation, such as setting the zero-reaching interval to -0.1 to 0.1, or -0.2 to 0.2, etc. There are no restrictions here. The formula for calculating the autocorrelation coefficient is a well-known constant for those skilled in the art, and it can be directly obtained through the built-in program of the data processing software. It will not be elaborated here. The method for optimizing the function expression of the trend drift fitting value is as follows: the function expression of the trend drift fitting value output by the trend drift fitting module is used as the initial function expression. When the scatter plot does not meet the randomness standard, the gradient descent method is used to update the polynomial coefficients in the initial function expression until the scatter plot meets the randomness standard. When optimizing the function expression, it is ensured that the optimized function expression meets the constraint that the coefficient of determination is within the preset range of the coefficient of determination. Please see Figures 5-6 This embodiment provides a scatter plot of the random error values ​​of 50 consecutive sampling points before stripping the trend drift fitted values ​​under changing scenarios, and a line plot of the random drift value distribution after stripping. Figure 5 and Figure 6 It can be seen that before removing the trend drift fitted values, the random error values ​​of the 50 consecutive sampling points show an increasing trend over time, exhibiting a certain time dependence (see details). Figure 5 After processing by this technical solution, the random drift values ​​at 50 consecutive sampling points exhibit a random distribution (see details). Figure 6 This meets the technical expectations of the present technical solution; The trend drift removal module generates a trend drift fitting value of the gyroscope to be calibrated in the pressure plate sensor device under the current sampling based on the optimized function expression, and subtracts the trend drift fitting value from the angular velocity measurement value under the current sampling to obtain the initial value of angular velocity calibration. The random drift smoothing module is used to analyze the trend drift fitting value, operating temperature and operating humidity of the gyroscope to be calibrated under the current sampling to determine the minimum length of the smoothing window. Based on this minimum length, the initial value of angular velocity calibration is processed by the window averaging method to obtain the final value of angular velocity calibration. Please see Figure 7 , Figure 7 The accuracy rate of the pressure plate sensor device in monitoring the opening and closing status of the pressure plate in 2024 after using the calibration system of this technical solution is given. Except for July, August, September and October, the monitoring accuracy rate of the other months is above 90%, and the highest can reach above 96%. It can be seen that the calibration system of this technical solution has good advantages in practical application. The minimum length of the smoothing window is obtained based on the smoothing accuracy analysis model. This model, built using the TensorFlow deep learning framework, includes an input layer that receives the trend drift fitted value, operating temperature, operating humidity, and the smoothing window length; multiple hidden layers that process the input data, with ReLU selected as the activation function; and an output layer that outputs the smoothing relative error. The training process for the smoothing accuracy analysis model is as follows: During the changing scenario test, several sample points were randomly selected, and the trend drift fitting value, working temperature, working humidity, angular velocity measurement value, and angular velocity true value corresponding to each sample point were recorded. For each sample point, a smoothing window of random length was set, and the window averaging method was used to obtain the final calibrated angular velocity value of the sample point. Then, the ratio of the final calibrated angular velocity value to the true angular velocity value of the same sample point was calculated as the smoothing accuracy of the sample point. Finally, the absolute difference between the smoothing accuracy and 1 was calculated as the smoothing relative error of the sample point. The larger the smoothing relative error, the worse the smoothing effect and the worse the accuracy of removing random drift values. It should be noted that the method for obtaining the final value of angular velocity calibration based on the window averaging method is as follows: calculate the initial value of angular velocity calibration of the sample sampling point and all sampling points in the previous smoothing window, and then perform mean averaging to obtain the final value of angular velocity calibration of the sample sampling point. A sample set is formed by integrating the trend drift fitted values, operating temperature, operating humidity, smoothing window length, and smoothing relative error of each sample point. The sample set is divided into a training set, a test set, and a validation set in a 70:15:15 ratio, and the batch size and training period are set. The trend drift fitted values, operating temperature, operating humidity, and smoothing window length in the training set are used as inputs, and the corresponding smoothing relative error is used as the output label to train the smoothing accuracy analysis model. The root mean square error is used as the loss function. During training, the model parameters are updated through backpropagation to minimize the loss function value. Specifically, the model parameters can be updated through optimization algorithms such as Adam and SGD. During training, the model hyperparameters (such as learning rate, batch size, number of hidden layers, etc.) are adjusted through the validation set to optimize the model performance. After the predetermined training period is reached, the test set is input into the smoothing accuracy analysis model for performance testing. If the deviation between the predicted value and the true value of the smoothing relative error output by the smoothing accuracy analysis model is less than 3%, the training is considered complete; otherwise, training is repeated. The logic for obtaining the minimum length of the smoothing window under the current sampling is as follows: input the trend drift fitting value, working temperature, working humidity and the minimum length of the smoothing window under the previous sampling into the smoothing accuracy analysis model after training, and keep the trend drift fitting value, working temperature and working humidity unchanged. With the model output smoothing relative deviation less than a preset threshold as a constraint, the minimum length of the smoothing window under the previous sampling is gradually adjusted until the minimum value is found as the minimum length of the smoothing window of the gyroscope to be calibrated under the current sampling. It should be noted that determining the minimum length of the smoothing window under the current sampling reduces the amount of calculation required to obtain the final value of angular velocity calibration based on the window averaging method, thereby reducing the time delay of online gyroscope calibration. Furthermore, obtaining the final value of angular velocity calibration under the current sampling based on the window averaging method is an existing technology, which has been discussed earlier and will not be repeated here. It should be noted that using the minimum length of the smooth window under the previous sampling as the benchmark value for adjustment reduces the difficulty of searching and determining the minimum length of the smooth window under the current sampling, improves search efficiency, and thus reduces the time delay of online gyroscope calibration. The method for obtaining the minimum length of the smooth window under the previous sampling is as follows: a benchmark value of the smooth window is set in the initial sampling stage of the gyroscope to be calibrated, and the minimum length of the smooth window is iteratively updated as sampling progresses. It should be noted that setting the constraint that the smooth relative deviation is less than the preset threshold ensures the accuracy of the final value of the angular velocity calibration obtained by the window averaging method. The specific value of this preset threshold shall be determined by the staff according to the actual situation, such as being set between 1% and 5%, and is not limited here.

[0035] Example 2: Please see Figures 2-4 This embodiment provides a pressure plate sensor device for detecting the opening and closing state of a pressure plate on a power grid. It includes a gyroscope to be calibrated using the aforementioned calibration system, a pressure plate sensor body, and a clamping mechanism, wherein: The pressure plate sensor body 1 includes a pressure plate sensor housing, a circuit board fixed inside the pressure plate sensor housing, an antenna, and a battery. The circuit board is used to acquire the opening and closing status of the pressure plate on the power grid, and the gyroscope to be calibrated is mounted on the circuit board. The antenna is used to transmit the opening and closing status of the pressure plate on the power grid to the outside world, and the battery is used to power the circuit board. The circuit board mainly consists of a conventional microcontroller, LORA wireless communication chip, gyroscope, RF circuit, buttons, crystal oscillator and ball switch, etc., to realize real-time monitoring of the pressure plate position. The back of the circuit board uses 8 contact pads, which can realize sensor program download, number setting, and wireless communication function debugging. Specifically, there are two buttons on the circuit board: a side button and a center button. The side button is located on the edge of the circuit board and can only be operated by removing the rear cover of the pressure plate sensor. Pressing and holding the side button will prevent the pressure plate sensor from entering sleep mode. It has a serial port function, which can be used to configure the sensor number. The middle button is installed in the middle of the circuit board, and there is a small hole on the side of the pressure plate sensor housing for the middle button. The middle button can be pressed with a thin needle. A short press will send a data packet to the aggregation node. Press and hold for 3 seconds. If you release the button when the red light is on, it will obtain the number (i.e., the wireless configuration number, which needs to be sent to the aggregation node first) from the aggregation node. Otherwise, if you do not release the button, continue to press and hold until the blue light is on. If you release the button when the blue light is on, the calibration operation will be performed. The calibration is complete when the light stops flashing. The calibration here is to remove the deterministic error of the gyroscope through conventional calibration. The calibration for random error is performed in real time. When using the above calibration system, the calibration is performed by a microcontroller. The temperature and humidity required for calibration can be obtained by the temperature and humidity sensor built into the pressure plate sensor, or by the temperature and humidity sensor installed in the environment around the pressure plate sensor. The working time is obtained by the timer built into the pressure plate sensor. Otherwise, if you do not release the button when the blue light is on, continue to press and hold until the red and blue lights are on together. Releasing the button will put the device into sleep mode. The clamping mechanism 2 is fixedly connected to the surface of the pressure plate sensor housing, and the pressure plate sensor housing and the pressure plate are detachably connected through the clamping mechanism.

[0036] As one implementation, the clamping mechanism consists of at least two thin plates fixedly connected to the surface of the pressure plate sensor housing, as detailed in the following document. Figures 2-3 By inserting a thin sheet into the slot of the pressure plate, a detachable connection is achieved between the pressure plate sensor housing and the pressure plate. This thin sheet clamping mechanism is suitable for spring-loaded pressure plates.

[0037] As another embodiment, the clamping mechanism is a recessed retainer with an opening, which is fixedly connected to the surface of the pressure plate sensor housing. See details... Figure 4 By inserting the pressure plate into the concave bracket, a detachable connection is achieved between the pressure plate sensor housing and the pressure plate. The clamping mechanism of this concave bracket structure is suitable for continuous pressure plates.

[0038] The above formulas are all dimensionless calculations. The formulas are derived from software simulations based on a large amount of collected data to obtain the most recent real-world results. The preset parameters in the formulas are set by those skilled in the art according to the actual situation.

[0039] The above embodiments can be implemented, in whole or in part, by software, hardware, firmware, or any other combination thereof. When implemented in software, the above embodiments can be implemented, in whole or in part, as a computer program product. Those skilled in the art will recognize that the units and algorithm steps of the various examples described in conjunction with the embodiments disclosed herein can be implemented by electronic hardware, or a combination of computer software and electronic hardware. Whether these functions are implemented in hardware or software depends on the specific application and design constraints of the technical solution.

[0040] The units described as separate components may or may not be physically separate. The components shown as units may or may not be physical units; they may be located in one place or distributed across multiple network units. Some or all of the units can be selected to achieve the purpose of this embodiment according to actual needs.

[0041] The above description is merely a specific embodiment of this application, but the scope of protection of this application is not limited thereto. Any changes or substitutions that can be easily conceived by those skilled in the art within the scope of the technology disclosed in this application should be included within the scope of protection of this application.

Claims

1. A calibration system for calibrating a gyroscope in a pressure plate sensor device, characterized in that, include: Gyroscope testing equipment is used to test gyroscopes to obtain angular velocity measurements of the gyroscopes under various test scenarios, including the actual angular velocity value, operating temperature, and operating humidity. The trend drift fitting module is used to analyze the measured and true angular velocity values ​​under the test scenario to determine the random error value of the angular velocity. It also incorporates the working temperature and humidity of the test scenario and solves the functional expression of the trend drift fitting value based on the gradient descent method. The random drift evaluation module is used to combine the functional expression of the trend drift fitting value and the random error value of the gyroscope in the changing scenario to construct a scatter plot of working time-random drift value, and optimize the functional expression of the trend drift fitting value based on the scatter plot until the scatter plot meets the randomness standard. The trend drift removal module generates a trend drift fitting value of the gyroscope to be calibrated in the pressure plate sensor device under the current sampling based on the optimized function expression, and subtracts the trend drift fitting value from the angular velocity measurement value under the current sampling to obtain the initial value of angular velocity calibration. The random drift smoothing module is used to analyze the trend drift fitting value, operating temperature and operating humidity of the gyroscope to be calibrated under the current sampling to determine the minimum length of the smoothing window. Based on this minimum length, the initial value of angular velocity calibration is processed by the window averaging method to obtain the final value of angular velocity calibration.

2. The calibration system according to claim 1, characterized in that, The gyroscope testing equipment consists of a precision rotating platform, a temperature and humidity control device, and data acquisition software, wherein: A precision rotary platform is used to provide stable rotational motion for detecting gyroscopes at different angular velocities; Temperature and humidity control device is used to provide stable temperature and humidity during the measurement process of gyroscope, so as to verify the working performance of gyroscope under different operating temperatures / humidities; Data acquisition software is used to record angular velocity measurements, actual angular velocity values, operating temperature, and operating humidity during the integrated testing process.

3. The calibration system according to claim 1, characterized in that, The functional expression for the trend drift fitted value is as follows: ； ； In the formula, Indicates the first The angular velocity measurement value obtained during the second sampling. Indicates the first The true value of the angular velocity corresponding to the next sample. Indicates the gyroscope at the 1st Random error values ​​generated during each sampling. Index of the number of samples; In the formula, Indicates the gyroscope at the 1st Trend drift fitted value at the second sampling time These represent the gyroscope at the [number]th [time]. The duration of each sampling affects the fitted value, as do temperature and humidity.

4. The calibration system according to claim 3, characterized in that, The method for fitting and solving the functional expression of the trend drift fitted value is as follows: 1) Set up a standard test scenario, select an unused gyroscope and test it in the standard test scenario to obtain the angular velocity measurement value of the gyroscope in the standard test scenario; 2) Calculate the random error value of the gyroscope at each sampling time in the standard test scenario, and record the working time of the gyroscope at each sampling time. Construct a scatter plot of working time-random error value in the standard test scenario. Analyze the scatter plot based on data processing software. Use the determination coefficient being in the preset range of the determination coefficient as a constraint condition, and solve the functional expression of the influence of time on the fitted value by gradient descent method. 3) Obtain the angular velocity measurement value of the gyroscope under the standard humidity test scenario, and obtain the corresponding random error value and working time. Construct a scatter plot of working time-random error value under the standard humidity test scenario. Analyze the scatter plot based on data processing software. Use the determination coefficient being within the preset range of the determination coefficient as a constraint condition. Solve the functional expression of the temperature influence fitting value through the gradient descent method. The standard humidity test scenario is a test scenario with the same working humidity as the standard test scenario. 4) Obtain the angular velocity measurement value of the gyroscope in the test scenario, and obtain the corresponding random error value and working time. Construct a scatter plot of working time-random error value in the test scenario. Analyze the scatter plot based on data processing software. Use the determination coefficient being in the preset range of the determination coefficient as a constraint condition, and solve the functional expression of the humidity effect on the fitted value by gradient descent method.

5. The calibration system according to claim 1, characterized in that, The random drift value is the difference between the random error value and the trend drift fitted value; The condition for a scatter plot to satisfy the randomness criterion is that, within any working time period, the autocorrelation coefficient of the random drift value is within a preset zero-reaching interval for all lag periods. The method for optimizing the function expression of the trend drift fitting value is as follows: the function expression of the trend drift fitting value output by the trend drift fitting module is used as the initial function expression. When the scatter plot does not meet the randomness standard, the gradient descent method is used to update the initial function expression until the scatter plot meets the randomness standard. When optimizing the function expression, it is ensured that the optimized function expression meets the constraint that the coefficient of determination is within the preset range of the coefficient of determination.

6. The calibration system according to claim 1, characterized in that: The minimum length of the smoothing window is obtained based on the smoothing accuracy analysis model. The smoothing accuracy analysis model built on the TensorFlow deep learning framework includes an input layer for receiving trend drift fitting values, working temperature, working humidity and smoothing window length, multiple hidden layers for data processing of the input data, with ReLU selected as the activation function on the hidden layers, and an output layer for outputting the smoothing relative error. The calculation logic for the smoothing relative error is as follows: the window averaging method is used to obtain the final value of the angular velocity calibration, then the ratio of the final value of the angular velocity calibration to the true value of the angular velocity is calculated as the smoothing accuracy, and finally the absolute difference between the smoothing accuracy and 1 is calculated as the smoothing relative error.

7. The calibration system according to claim 6, characterized in that, The logic for obtaining the minimum length of the smoothing window under the current sampling is as follows: Input the trend drift fitting value, working temperature, working humidity and the minimum length of the smoothing window under the previous sampling under the current sampling into the smoothing accuracy analysis model after training, and keep the trend drift fitting value, working temperature and working humidity unchanged. Use the smoothing relative deviation output by the smoothing accuracy analysis model being less than a preset threshold as a constraint condition, and gradually adjust the minimum length of the smoothing window under the previous sampling until the minimum value is found as the minimum length of the smoothing window of the gyroscope to be calibrated under the current sampling.

8. A pressure plate sensor device for detecting the open / closed state of a pressure plate on a power grid, comprising a gyroscope to be calibrated using the calibration system of any one of claims 1-7, characterized in that... It also includes the pressure plate sensor body and the clamping mechanism, wherein: The pressure plate sensor body includes a pressure plate sensor housing, a circuit board fixed inside the pressure plate sensor housing, an antenna, and a battery. The circuit board is used to acquire the opening and closing state of the pressure plate on the power grid, and the gyroscope to be calibrated is mounted on the circuit board. The antenna is used to transmit the opening and closing state of the pressure plate on the power grid to the outside world, and the battery is used to power the circuit board. A clamping mechanism is fixedly connected to the surface of the pressure plate sensor housing, and the pressure plate sensor housing and the pressure plate are detachably connected through the clamping mechanism.

9. The pressure plate sensor device according to claim 8, characterized in that: The clamping mechanism consists of at least two thin plates fixedly connected to the surface of the pressure plate sensor housing. By inserting the thin plates into the slots of the pressure plate, a detachable connection between the pressure plate sensor housing and the pressure plate is achieved.

10. The pressure plate sensor device according to claim 8, characterized in that: The clamping mechanism is a recessed bracket with an opening that is fixedly connected to the surface of the pressure plate sensor housing. By inserting the pressure plate into the recessed bracket, a detachable connection between the pressure plate sensor housing and the pressure plate is achieved.

Images