A method for measuring the height of hydraulic support based on magnetostrictive displacement measurement

By installing magnetostrictive displacement sensors and other sensors in the column cylinder of the hydraulic support, combined with data processing algorithms, the problem of large error in the height measurement of hydraulic support is solved, and high-precision and automated hydraulic support height measurement is achieved.

CN118776446BActive Publication Date: 2024-11-08GUANGDONG RUNYU SENSOR CO LTD
View PDF 0 Cites 0 Cited by

Patent Information

Application Number
CN202411278942.6
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2024-09-12
Publication Date
2024-11-08
Estimated Expiration
2044-09-12

AI Technical Summary

Technical Problem

The existing hydraulic support height measurement methods have problems such as large measurement errors, manual participation and difficulty in meeting automation requirements in underground coal mine environments.

Method used

Using a magnetostrictive displacement sensor based method, by installing the first and second magnetostrictive displacement sensors in the column cylinder, combining the inclination sensor, temperature sensor and inertia measurement unit, data acquisition and error compensation are performed, and Kalman filtering algorithm is used. and the least squares support vector machine model perform data processing and calculate the height of the hydraulic bracket.

Benefits of technology

Real-time and accurate measurement of hydraulic bracket height is achieved, environmental interference errors are reduced, measurement accuracy and stability are improved, and the needs of manual participation and regular maintenance are avoided.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN118776446B_ABST
    Figure CN118776446B_ABST
Patent Text Reader

Abstract

The present invention relates to a method for measuring the height of a hydraulic support based on magnetostrictive displacement measurement, and belongs to the field of data processing technology. The method comprises the following steps: installing a magnetostrictive displacement sensor according to the structure of the hydraulic support column cylinder; collecting inclination data, temperature data and vibration data through a sensor arranged in an electronic compartment; preprocessing the collected state data; compensating the displacement data obtained by the magnetostrictive displacement sensor based on the decomposition values ​​of the temperature data and vibration data collected by the sensor through an error compensation model; and calculating the height of the hydraulic support using trigonometric functions through the obtained inclination fusion value and displacement compensation value. The present invention arranges a magnetostrictive displacement sensor in the column cylinder to collect data on the real-time state of the column cylinder, and processes the measurement data using error compensation, thereby eliminating interference errors caused by the working environment of the hydraulic support.
Need to check novelty before this filing date? Find Prior Art

Description

Technical Field

[0001] The invention belongs to the technical field of data processing, and in particular relates to a method for measuring the height of a hydraulic support based on magnetostrictive displacement measurement. Background Art

[0002] As one of the important electromechanical equipment of intelligent working face, hydraulic support not only plays an important role in supporting the roof of the mining area, but also provides a safe working space for coal mining machines, scraper conveyors and underground personnel. At present, the automation level of hydraulic support working faces in coal mines is constantly improving. In the actual production process, the hydraulic support may deviate from the original ideal posture due to the ups and downs of the roof and floor of the working face, changes in coal seam thickness, periodic pressure, and impact ground pressure, which seriously threatens the safe and efficient production of the working face. Therefore, the real-time and accurate measurement of the height data of the hydraulic support is becoming more and more important.

[0003] Currently, height measuring rods, laser distance measurement, inclination sensor calculation and other methods are used. The height measuring rod is hung on the outside of the support column, and the height is read and recorded manually; the laser distance measurement has high measurement accuracy, but it is affected by the coal dust environment, the laser emission window lens is easily covered by coal dust, and the target point is not fixed, so the measurement error is large; the above methods require manual participation or frequent maintenance, which is difficult to meet the requirements of hydraulic support operation.

[0004] Therefore, it is urgent to propose a hydraulic support height measurement method based on magnetostrictive displacement measurement. By setting a magnetostrictive displacement sensor in the column cylinder, real-time data acquisition is performed, and the measurement data is calibrated by error compensation. Summary of the invention

[0005] In order to solve the above problems existing in the prior art, the present invention provides a method for measuring the height of a hydraulic support based on magnetostrictive displacement measurement. A magnetostrictive displacement sensor is arranged in the column cylinder to collect data on the real-time status of the column cylinder, and the measurement data is processed by error compensation, thereby eliminating the interference error caused by the working environment of the hydraulic support.

[0006] The purpose of the present invention can be achieved through the following technical solutions:

[0007] The present disclosure provides a method for measuring the height of a hydraulic support based on magnetostrictive displacement measurement, comprising the following steps:

[0008] S1. Install the displacement sensor: According to the structure of the hydraulic support column cylinder, the first magnetostrictive displacement sensor is coaxially installed inside the cylinder barrel to measure the displacement of the primary piston; the second magnetostrictive displacement sensor is installed on one side of the central axis of the secondary piston rod to measure the displacement of the secondary piston;

[0009] S2, obtaining status data: collecting inclination data, temperature data and vibration data through the inclination sensor, temperature sensor and inertial measurement unit arranged in the electronic compartment of the magnetostrictive displacement sensor, and sending the data to the data processing unit through the Bluetooth signal transceiver;

[0010] S3, data preprocessing: the data processing unit preprocesses the received state data, performs temperature-based weighted averaging on the inclination data and fuses it with the attitude angle data obtained by the inertial measurement unit using the Kalman filter algorithm, and decomposes the vibration data through the inclination fusion value obtained by the fusion to obtain the vibration value perpendicular to the inclination direction;

[0011] S4, obtaining displacement data: using temperature data, vibration data, displacement data before compensation and the resonant frequency of the waveguide wire as input and displacement data after compensation as output, to establish an error compensation model, wherein the displacement data before compensation is obtained by a magnetostrictive displacement sensor;

[0012] The displacement data acquired by the magnetostrictive displacement sensor is compensated by the error compensation model based on the temperature data and vibration data decomposition values ​​collected by the sensor to obtain the displacement data of the column cylinder piston;

[0013] S5. Calculate the support height: Calculate the height of the hydraulic support using trigonometric functions through the obtained inclination fusion value and displacement compensation value.

[0014] Furthermore, the data preprocessing step includes the following sub-steps:

[0015] S31, calculating inclination data: calculating the inclination correction value of the column cylinder according to the temperature compensation model through the temperature data and inclination data respectively obtained by the first magnetostrictive displacement sensor and the second magnetostrictive displacement sensor;

[0016] S32, attitude angle fusion: the attitude angle data of the column cylinder is measured in real time by the inertial measurement unit, the attitude angle data is filtered by the Kalman filter algorithm, and the state of a system is estimated by an underestimation method, and the inclination angle fusion value is obtained by fusing the attitude angle data and the inclination correction value;

[0017] S33, obtaining vibration data: obtaining vibration data through an inertial measurement unit, decomposing the vibration data according to the inclination fusion value, and obtaining a vibration value perpendicular to the inclination direction, wherein the vibration value includes vibration intensity and vibration direction.

[0018] Furthermore, the temperature compensation model is expressed as:

[0019] ;

[0020] In the formula, is the inclination correction value, and are the inclination values ​​measured by the inclination sensors in the first magnetostrictive displacement sensor and the second magnetostrictive displacement sensor, respectively. and are the temperature values ​​measured by the temperature sensors of the first magnetostrictive displacement sensor and the second magnetostrictive displacement sensor, respectively, and are reference temperatures of the inclination sensors corresponding to the first magnetostrictive displacement sensor and the second magnetostrictive displacement sensor respectively; and are the coefficients of the polynomial, N is the order of the polynomial, and is the weight coefficient, which is allocated according to the temperature influence coefficient of the inclination sensor.

[0021] Furthermore, the posture angle fusion step includes the following sub-steps:

[0022] S321, System state model: define the state vector M , including the state variables of the inclination angle, the inclination angle and the angular velocity, expressed as:

[0023] ;

[0024] S322, System Dynamics Model: Under the assumption of a linear system, using the state transfer matrix A and the input matrix B Represents the system's state update process, describing the process of how the system state evolves over time m k :

[0025] m k = A * m k-1 + B * u k + w k ;

[0026] In the formula, u k For external input, w k is the process noise;

[0027] S323. Observation model: Using the observation matrix H and observation noise v k Characterization Measurement Values z k , define an observation equation:

[0028] z k = H * m k + v k ;

[0029] S324, Kalman filter: at every moment k , use the system dynamic model to predict the estimated value and covariance matrix of the current state; combine the observation model and sensor measurement, and use the Kalman gain to correct the predicted value to obtain the final state estimate and covariance matrix; by comparing the measured data with the predicted value, the Kalman gain calculation adjusts the predicted value.

[0030] Furthermore, the step of obtaining displacement data includes the following sub-steps:

[0031] S41, establishing a data set: applying vibration interference perpendicular to the measuring rod under different temperature environments, measuring the measurement data of the magnetostrictive displacement sensor, and calibrating it as label data through high-precision measuring equipment to establish a data set;

[0032] S42. Data processing: clean the data set, remove or correct missing values ​​and outliers, extract relevant features from the original data, and normalize the features;

[0033] S43, constructing an error compensation model: taking the temperature data, the vibration data, the displacement data before compensation and the resonant frequency of the waveguide as inputs and the displacement data after compensation as outputs, to establish an error compensation model;

[0034] S44, model training: inputting the optimized hyperparameters into the error compensation model, using the data set to train and evaluate the model, determining the fitness of the current hyperparameters, and adjusting the model parameters to minimize the error;

[0035] S45, displacement data compensation: input the collected and processed real-time temperature data, vibration data, displacement data before compensation, and the waveguide wire resonant frequency of the magnetostrictive displacement sensor into the error compensation model to obtain the compensated displacement data.

[0036] Furthermore, the step of constructing the error compensation model includes the following sub-steps:

[0037] S431, establish a least squares support vector machine model, the regression function is:

[0038] ;

[0039] In the formula, l is the number of samples, xi For thei The input of the iteration, x is the input data set, b is the bias term, αi is the Lagrange multiplier, k ( x , xi ) is the kernel function that satisfies the Mercer condition;

[0040] Select the kernel function, where the kernel function K The expression is:

[0041] K = aK poly + (1- a ) K RBF ;

[0042] In the formula, K poly is the polynomial kernel function, K RBF is the RBF kernel function, a is the mixed weight coefficient of the polynomial kernel function and the RBF kernel function;

[0043] S432, the RBF kernel function parameters in the hybrid function are optimized by particle swarm optimization algorithm γ Value, penalty coefficient c Values, polynomial kernel function parameters q Values ​​and the mixed weight coefficients of the two kernel functions a Make optimizations.

[0044] Furthermore, the step of constructing the error compensation model further includes the following sub-steps:

[0045] S4321, set the particle swarm size, and randomly generate a group of ( γ , c , q , a ) as the initial position of the particle; determine the value range of the parameter to be optimized and set the maximum speed;

[0046] S4322, fitness value of each particle in the particle swarm F ( x i ) is used for calculation, the particle swarm is divided into several subgroups, denoted as S, the optimal particle with the best fitness value in the S subgroups is calculated, the optimal solution of each subgroup in the S subgroups is obtained, and the particle with the best fitness value in the subgroup is selected to form a population gene library;

[0047] Among them, the expression of the fitness value function is:

[0048] ;

[0049] In the formula, yi For the i The measured value of samples; fi For the i The predicted value of samples, m is the total size of particle swarm;

[0050] S4323, calculating the shrinkage and expansion coefficients and the learning probability parameters of the subgroup, and iteratively updating the fitness value of the particles and the optimal solution of the population;

[0051] S4324. When the iteration reaches the evolution cycle, the inferior particles in the sub-group are eliminated and the gene pool of the established population is updated;

[0052] S4325. Return to the step of learning the probability parameters of the contraction-expansion coefficient and the subgroup, repeat the above steps until the iteration is completed, solve the optimal solution of the population, and obtain the optimal value of the hyperparameter.

[0053] Furthermore, the column oil cylinder comprises a cylinder barrel, a primary piston and a primary piston rod are movably arranged in the cylinder barrel, a secondary piston and a secondary piston rod are movably arranged in the primary piston rod; the primary piston and the primary piston rod are integrally connected, and the secondary piston and the secondary piston rod are integrally connected;

[0054] The first magnetostrictive displacement sensor and the second magnetostrictive displacement sensor both include a waveguide wire, a measuring rod, an electronic compartment and a magnetic ring; each of the electronic compartments is provided with an inclination sensor, a temperature sensor and an inertial measurement unit;

[0055] The electronic compartment of the first magnetostrictive displacement sensor is embedded in the cylinder barrel, the measuring rod of the first magnetostrictive displacement sensor passes through the primary piston and extends into the secondary piston rod, and the magnetic ring 1 of the first magnetostrictive displacement sensor is embedded in the end surface center of the primary piston; the electronic compartment of the second magnetostrictive displacement sensor is embedded in the end of the secondary piston rod, the measuring rod of the second magnetostrictive displacement sensor is installed on one side of the secondary piston rod and extends inward, and the magnetic ring 2 of the second magnetostrictive displacement sensor is embedded in the end surface center of the primary piston rod and close to one side of the measuring rod of the second magnetostrictive displacement sensor; the magnetic ring 1 of the first magnetostrictive displacement sensor is annular, and the magnetic ring 2 of the second magnetostrictive displacement sensor is an open semi-annular.

[0056] Furthermore, the inclination sensor is used to measure the inclination angle of the column cylinder; the temperature sensor is used to measure the temperature of the first magnetostrictive displacement sensor and the second magnetostrictive displacement sensor; and the inertial measurement unit is used to measure the posture angle data of the column cylinder and the intensity and direction of mechanical vibration.

[0057] Furthermore, in the step of calculating the support height, the height calculation formula of the hydraulic support is:

[0058] ;

[0059] In the formula, H is the height of the hydraulic support, L is the total length of the column cylinder after displacement, A It is the inclination fusion value of the column cylinder.

[0060] The beneficial effects of the present invention are:

[0061] The present invention firstly arranges the magnetostrictive displacement sensor in a column oil cylinder based on the characteristics of non-contact continuous measurement, long service life, high precision, high repeatability measurement, anti-radio frequency interference, and no need for regular calibration and maintenance. The interference problem between the displacement sensors is avoided through the designed installation position. Based on the installed displacement sensor, the temperature sensor, the inclination sensor and the inertial measurement unit are arranged in the electronic compartment of the displacement sensor to collect state data; then, based on the conditions of the working environment of the hydraulic support, the environmental data is collected to correct and compensate the key parameters for calculating the height of the hydraulic support: the inclination angle of the column oil cylinder and the displacement of the piston in the column oil cylinder, so as to eliminate environmental interference and reduce measurement errors.

[0062] The present invention includes the following in the process of correcting the inclination angle of the column cylinder: based on the two inclination sensors set up in the structure, the weights of the two sensors in the angle value calculation are allocated through the temperature influence coefficient of the sensor itself, so as to eliminate temperature interference and sensor error; based on the inclination correction value and the column cylinder attitude angle data measured by the inertial measurement unit, data fusion is performed to comprehensively utilize the advantages of both and improve the overall accuracy and precision of attitude angle measurement. The process of displacement compensation includes: based on the main external influencing factors of the magnetostrictive displacement sensor error, the temperature data and vibration data are measured, and the vibration data is decomposed through the inclination data finally obtained to obtain the direct parameters of the forced vibration inside the sensor, and finally the measured data and the data of the sensor itself are used as input to construct an error compensation model, obtain displacement compensation data, and reduce the influence of environmental interference in the application scenario on the sensor accuracy. BRIEF DESCRIPTION OF THE DRAWINGS

[0063] In order to facilitate understanding by those skilled in the art, the present invention is further described below with reference to the accompanying drawings.

[0064] Figure 1 A schematic diagram of the structure of a hydraulic support provided by an embodiment of the present invention;

[0065] Figure 2 A schematic diagram of the structure of a column cylinder provided in an embodiment of the present invention;

[0066] Figure 3 A schematic diagram of installing a first magnetostrictive displacement sensor provided in an embodiment of the present invention;

[0067] Figure 4 A schematic diagram of the installation of a second magnetostrictive displacement sensor provided in an embodiment of the present invention;

[0068] Figure 5 A schematic diagram of the structure of a displacement sensor provided by an embodiment of the present invention;

[0069] Figure 6 A schematic diagram of the steps of a method for measuring the height of a hydraulic support based on magnetostrictive displacement measurement provided by an embodiment of the present invention;

[0070] Figure 7 A schematic diagram of the steps of data preprocessing provided by an embodiment of the present invention;

[0071] Figure 8 A schematic diagram of the steps for obtaining displacement data provided by an embodiment of the present invention;

[0072] Description of main component symbols:

[0073] In the figure: 1. top beam; 2. shielding beam; 3. column cylinder; 4. base; 5. cylinder barrel; 6. primary piston; 7. primary piston rod; 8. secondary piston; 9. secondary piston rod; 10. first magnetostrictive displacement sensor; 11. magnetic ring one; 12. magnetic ring two; 13. second magnetostrictive displacement sensor; 14. electronic compartment; 15. measuring rod. DETAILED DESCRIPTION

[0074] In order to further explain the technical means and effects adopted by the present invention to achieve the predetermined invention purpose, the specific implementation methods, structures, features and effects of the present invention are described in detail below in conjunction with the accompanying drawings and preferred embodiments.

[0075] The structural composition of the hydraulic support is as follows: Figure 1 As shown, it mainly includes the following parts: top beam 1, shielding beam 2, base 4 and column cylinder 3, etc., among which the column cylinder 3 is the power core of the hydraulic support, mainly including a hydraulic cylinder body, a piston, a piston rod and a sealing structure, which is responsible for generating hydraulic power to realize the lifting and supporting function of the support.

[0076] The hydraulic support height measurement method proposed in the present application uses a built-in magnetostrictive displacement sensor and an inclination sensor to measure the telescopic displacement of the hydraulic support column cylinder 3 and the inclination angle of the column cylinder 3, and calculates the actual height of the hydraulic support based on trigonometric functions.

[0077] Specifically, each step of the method for measuring the height of a hydraulic support based on magnetostrictive displacement measurement of the present invention is described in detail through the following contents:

[0078] A method for measuring the height of a hydraulic support based on magnetostrictive displacement measurement, such as Figure 6 As shown, the following steps are included:

[0079] S1. Install the displacement sensor: Figure 2-4 As shown, according to the structure of the hydraulic support column cylinder 3, the first magnetostrictive displacement sensor 10 is coaxially installed inside the cylinder 5, and the magnetic ring 11 is installed at the bottom of the first-stage piston 6 to measure the displacement of the first-stage piston 6;

[0080] The second magnetostrictive displacement sensor 13 is installed on one side of the central axis of the secondary piston rod 9 , and the second magnetic ring 12 is installed on the top of the primary piston rod 7 to measure the displacement of the secondary piston 8 .

[0081] It should be noted that in this embodiment, the hydraulic support column cylinder 3 is a two-stage cylinder and needs to be installed with two displacement sensors. If two displacement sensors are installed in one cylinder at the same time, there will be interference. Therefore, the two displacement sensors should be installed in the two cylinders of the column respectively.

[0082] Specifically, the column cylinder 3 includes a cylinder barrel 5, in which a primary piston 6 and a primary piston rod 7 are movably arranged, and in which a secondary piston 8 and a secondary piston rod 9 are movably arranged. The primary piston 6 and the primary piston rod 7 are integrally connected, and the secondary piston 8 and the secondary piston rod 9 are integrally connected; a first magnetostrictive displacement sensor 10 is installed inside the cylinder barrel 5, and the cylinder barrel 5 is provided with a space for the sensor to be wired out. The wire-out space is sealed with the cylinder barrel 5 through the side seal of the O-ring and the retaining ring, and is used to measure the displacement of the primary piston 6; a second magnetostrictive displacement sensor 13 is installed in the secondary piston rod 9, and is used to measure the displacement of the secondary piston 8.

[0083] The first magnetostrictive displacement sensor 10 and the second magnetostrictive displacement sensor 13 both include a magnetostrictive wire (hereinafter referred to as a waveguide wire), a measuring rod 15, an electronic compartment 14, and a non-contact magnetic ring (with a permanent magnet installed therein) sleeved on the measuring rod 15. The electronic compartment 14 of the first magnetostrictive displacement sensor 10 is embedded in the cylinder 5, the measuring rod 15 of the first magnetostrictive displacement sensor 10 passes through the primary piston 6 and extends into the secondary piston rod 9, and the magnetic ring 11 of the first magnetostrictive displacement sensor 10 is embedded in the end face center of the primary piston 6; the electronic compartment 14 of the second magnetostrictive displacement sensor 13 is embedded in the end of the secondary piston rod 9, the measuring rod 15 of the second magnetostrictive displacement sensor 13 is installed on one side of the secondary piston rod 9 and extends inward, and the magnetic ring 2 12 is embedded in the end face center of the primary piston rod 7 and close to one side of the measuring rod 15 of the second magnetostrictive displacement sensor 13. The magnetic ring 11 of the first magnetostrictive displacement sensor 10 is annular and slidably mounted on the outer surface of the measuring rod 15 thereof, and the magnetic ring 2 12 of the second magnetostrictive displacement sensor 13 is in the shape of an open half ring.

[0084] Each electronic compartment 14 is provided with an inclination sensor, a temperature sensor and an inertial measurement unit. The inclination sensor is used to measure the inclination angle of the column cylinder 3; the temperature sensor is used to measure the temperature of the first magnetostrictive displacement sensor 10 and the second magnetostrictive displacement sensor 13; the inertial measurement unit is used to measure the intensity and direction of mechanical vibration. It is understandable that the inertial measurement unit includes an accelerometer, a gyroscope and a magnetometer, etc., so as to provide comprehensive vibration data: the intensity, direction and possible rotation of the vibration.

[0085] The circuit design of the magnetostrictive displacement sensor mainly includes: power protection circuit, Bluetooth module, battery boost and control circuit, battery voltage acquisition circuit, EEPROM circuit, built-in antenna circuit and anti-reverse connection circuit. A disposable large-capacity battery is used as the power supply to power the system main circuit and boost circuit respectively. The whole machine is powered on by the Hall switch, and the Hall switch pin level is detected to determine whether it needs to be shut down and replaced by the battery.

[0086] S2, obtaining status data: collecting inclination data, temperature data and vibration data through sensors arranged in the electronic compartment 14, and sending the data to the data processing unit through the Bluetooth signal transceiver.

[0087] It can be understood that the displacement sensor is based on Bluetooth wireless transmission, and a Bluetooth signal transceiver with a built-in removable battery is set on the outside. The displacement sensor is connected to the Bluetooth signal transceiver through a mining connector to realize Bluetooth wireless communication function.

[0088] S3, data preprocessing: The data processing unit preprocesses the received status data to obtain the inclination data and vibration data of the column cylinder 3, such as Figure 7 As shown, the following steps are included:

[0089] S31, calculating the inclination data: calculating the inclination correction value of the column cylinder 3 according to the temperature data and the inclination data respectively obtained by the first magnetostrictive displacement sensor 10 and the second magnetostrictive displacement sensor 13 according to the temperature compensation model, wherein the temperature compensation model is:

[0090] ;

[0091] In the formula, is the inclination correction value, and are the inclination values ​​measured by the inclination sensors in the first magnetostrictive displacement sensor 10 and the second magnetostrictive displacement sensor 13, respectively. and are the temperature values ​​measured by the temperature sensors of the first magnetostrictive displacement sensor 10 and the second magnetostrictive displacement sensor 13, respectively, and are the reference temperatures of the inclination sensors corresponding to the first magnetostrictive displacement sensor 10 and the second magnetostrictive displacement sensor 13 respectively; and are the coefficients of the polynomial, N is the order of the polynomial, and is the weight coefficient, which is allocated according to the temperature influence coefficient obtained from the technical manual of the inclination sensor.

[0092] It can be understood that in this embodiment, the data measured by two existing inclination sensors are corrected by the corresponding working environment temperature, the weights of the two sensors in the angle value calculation are allocated according to the temperature influence coefficient of the sensor itself, and the temperature compensation curve is adjusted more finely based on polynomial fitting to better match the actual temperature characteristics of the sensor, thereby ensuring that more accurate inclination data is obtained under different ambient temperatures.

[0093] S32, attitude angle fusion: the attitude angle data of the column cylinder 3 is measured in real time by the inertial measurement unit, the attitude angle data is filtered by the Kalman filter algorithm, and the state of a system is estimated by an underestimation method, and the inclination angle fusion value is obtained by information fusion of the attitude angle data and the inclination correction value, including the following steps:

[0094] S321, System state model: define the state vector M , including the state variables of the inclination angle, the inclination angle and the angular velocity, expressed as:

[0095] ;

[0096] S322, System Dynamics Model: Under the assumption of a linear system, using the state transfer matrix A and the input matrix B Represents the system's state update process, describing the process of how the system state evolves over time m k :

[0097] m k = A * m k-1 + B * u k + w k ;

[0098] In the formula, u k For external input, w k is the process noise.

[0099] S323, Observation model: define an observation equation, using the observation matrix H and observation noise v k Characterization Measurement Values z k , which represents how to obtain observations of the system state from sensor measurements:

[0100] z k = H * m k + v k ;

[0101] S324, Kalman filter: at every moment k , using the system dynamic model to predict the estimated value and covariance matrix of the current state; during the prediction process, the Kalman filter will deduce the state according to the dynamic behavior of the system, and reflect the uncertainty of the state estimation through the covariance matrix, so that the predicted state value is reliable and accurate;

[0102] Combining the observation model and sensor measurements, the predicted value is corrected through the Kalman gain to obtain the final state estimate and covariance matrix; by comparing the measured data with the predicted value, the Kalman gain calculation can adjust the predicted value to more accurately reflect the actual state of the system.

[0103] By fusing the attitude angle data and the inclination correction value, a more accurate state estimation is provided. By modeling the system dynamics and assuming the Gaussian distribution of noise, the Kalman filter can effectively combine the sensor data, overcome the limitations of the inclination sensor and the gyroscope, and accurately estimate the system state.

[0104] S33, obtaining vibration data: obtaining vibration data through an inertial measurement unit, decomposing the vibration data according to the inclination fusion value, and obtaining a vibration value perpendicular to the inclination direction, wherein the vibration value includes vibration intensity and vibration direction.

[0105] It is understandable that the inertial measurement unit can measure the acceleration of an object in the three axes (x, y, z) through the built-in acceleration sensor, and calculate the changes in velocity and displacement by integrating the acceleration signal to determine the intensity of the vibration. The angular velocity or angle change rate of the object in the three axes (x, y, z) is measured by the gyroscope sensor to determine the direction of the vibration.

[0106] It should be noted that the inclination sensor is easily affected by the change of ambient temperature, resulting in measurement errors, namely, temperature drift. To address this problem, this embodiment, based on the displacement sensor installation structure and the sensor set inside the electronic compartment 14, comprehensively considers the correction of the sensor measurement value by temperature and the weighted average of the sensor measurement values ​​at different temperatures, designs a temperature compensation model, and applies it to actual measurement projects to compensate for the temperature drift error generated by the inclination sensor, effectively improving the stability and accuracy of the inclination sensor.

[0107] In addition, the vibration data is decomposed based on the inclination data to obtain the vibration value perpendicular to the direction of the column cylinder 3. The form of mechanical vibration is taken into account in the interference of the magnetostrictive displacement sensor. The vibration state of the main direction is decomposed to cope with different forms of vibration interference, thereby removing interference and improving displacement measurement data in a more practical scenario.

[0108] Since the magnetostrictive displacement sensor in this embodiment is applied to the hydraulic support, the external impact will cause the waveguide wire of the magnetostrictive displacement sensor to vibrate, resulting in vertical displacement. The vibration interference of the waveguide wire affects the length of the waveguide wire, the magnetostrictive coefficient of the waveguide wire, and the spiral magnetic field of the waveguide wire. These parameters are closely related to the output voltage of the magnetostrictive displacement sensor, so the vibration interference will affect the output voltage of the magnetostrictive displacement sensor. Therefore, in this embodiment, the vibration state of the hydraulic support is monitored, and the measured displacement data is compensated for errors according to the intensity and frequency of the vibration.

[0109] S4. Obtaining displacement data: Compensating the displacement data obtained by the magnetostrictive displacement sensor based on the temperature data and vibration data through the error compensation model, such as Figure 8 As shown, the following steps are included:

[0110] S41, establishing a data set: applying vibration interference perpendicular to the measuring rod 15 under different temperature environments, measuring the measurement data of the magnetostrictive displacement sensor, and calibrating it as label data through high-precision measuring equipment to establish a data set.

[0111] S42. Data processing: Clean the data set, remove or correct missing values ​​and outliers, extract relevant features from the original data, and normalize the features.

[0112] S43, constructing an error compensation model: taking the temperature data, the vibration data, the displacement data before compensation and the resonant frequency of the waveguide as inputs and the displacement data after compensation as outputs, and establishing an error compensation model, including the following steps:

[0113] S431. Establish a least squares support vector machine (LS-SVM) model, and the regression function is:

[0114] ;

[0115] In the formula, l is the number of samples, xi For the i The input of the iteration, x is the input data set, b is the bias term, αi is the Lagrange multiplier, k ( x , xi ) is the kernel function that satisfies the Mercer condition.

[0116] Select the kernel function, where the kernel function K The expression is:

[0117] K = aK poly + (1- a ) K RBF ;

[0118] In the formula, K poly is the polynomial kernel function, K RBF is the RBF kernel function, a is the mixed weight coefficient of the polynomial kernel function and the RBF kernel function.

[0119] LS-SVM combines the advantages of neural network and support vector machine, not only has fast training speed but also has strong generalization ability. In this embodiment, LS-SVM is selected to construct the error compensation model.

[0120] S432, the RBF kernel function parameters in the hybrid function are optimized by particle swarm optimization algorithm γ Value, penalty coefficient c Values, polynomial kernel function parameters q Values ​​and the mixed weight coefficients of the two kernel functions a The optimization includes the following steps:

[0121] S4321, set the particle swarm size, and randomly generate a group of ( γ , c , q , a ) as the initial position of the particle; determine the value range of the parameter to be optimized and set the maximum speed.

[0122] Penalty Factor c reflects the trade-off between model complexity and approximation error, c The larger it is, the better the fit; It reflects the strength of the connection between the support vectors. After multiple trials, it is finally determined that ( γ , c , q , a )scope.

[0123] S4322, fitness value of each particle in the particle swarm F ( x i ) is used for calculation, the particle swarm is divided into several subgroups, denoted as S, the optimal particle with the best fitness value in the S subgroups is calculated, the optimal solution of each subgroup in the S subgroups is obtained, and the particle with the best fitness value in the subgroup is selected to form a population gene library;

[0124] Among them, the expression of the fitness value function is:

[0125] ;

[0126] In the formula, yi For the i The measured value of samples; fi For the i The predicted value of samples, m is the total size of particle swarm.

[0127] S4323, calculating the shrinkage and expansion coefficients and the learning probability parameters of the subgroup, and iteratively updating the fitness value of the particles and the optimal solution of the population;

[0128] S4324. When the iteration reaches the evolution cycle, the inferior particles in the sub-group are eliminated and the gene pool of the established population is updated;

[0129] S4325. Return to step S4323, repeat the above steps until the iteration is completed, solve the optimal solution of the population, and obtain the optimal value of the hyperparameter.

[0130] The particle swarm optimization of the traditional RBF kernel SVM must optimize two parameters (RBF kernel function parameters and penalty coefficient c). The LS-SVM parameter selection problem selected in this embodiment is: a combination of four parameters, namely, RBF kernel function parameters, penalty coefficient c, polynomial kernel function parameters q, and mixed weight coefficient a, so that the LS-SVM has the best performance. The PSO algorithm is used to optimize the parameters ( γ , c , q , a ) is optimized to obtain the error compensation model.

[0131] S44, model training: input the optimized hyperparameters into the error compensation model, use the data set to train and evaluate the model, determine the fitness of the current hyperparameters, and adjust the model parameters to minimize the error.

[0132] S45, displacement data compensation: input the collected real-time temperature data, vibration data, displacement data before compensation, and the waveguide wire resonant frequency of the magnetostrictive displacement sensor into the error compensation model to obtain the compensated displacement data.

[0133] It should be noted that, in this embodiment, temperature data, vibration data, displacement data before compensation and the resonant frequency of the waveguide wire are used as inputs, and displacement data after compensation is used as output, a least squares support vector machine is constructed, and a particle swarm optimization algorithm is used to obtain hyperparameters. Training and verification are performed through sample data sets to obtain an error compensation model, which is used to solve the displacement of the piston in the column cylinder 3 in the bracket height measurement. The environmental impact of the magnetostrictive sensor applied to the hydraulic bracket measurement is incorporated into the model training, and its main mechanism of action is introduced into the model, thereby establishing a more realistic compensation model and improving the prediction ability and reliability of the model.

[0134] S5. Calculate the support height: Calculate the height of the hydraulic support by obtaining the inclination fusion value and the displacement compensation value. The calculation formula is:

[0135] ;

[0136] In the formula, H is the height of the hydraulic support, L is the total length of the column cylinder 3 after displacement, A It is the inclination fusion value of column cylinder 3.

[0137] In the actual calculation process, the distance between the ground of the base 4 and the lower end of the column cylinder 3 is measured and recorded as L 3, which is a fixed value; and measure the distance between the top surface of the top beam 1 and the upper end of the column cylinder 3, and record it as L 4, which is a fixed value; and the displacement sensor is used to measure the displacement of the first-stage piston 6 in the column cylinder 3 and record it as L 1; and the displacement sensor is used to measure the displacement of the secondary piston 8 in the column cylinder 3 and record it as L 2, therefore, the total length of the column cylinder 3 after displacement L = L 1+ L 2+ L 3+ L 4.

[0138] It should be noted that the above calculation formula is calculated assuming that the support surface of the hydraulic support base 4 is a horizontal plane. When the hydraulic support has a slope, it is also necessary to make adaptive adjustments based on the above calculation according to the angle of the slope to obtain the height of the hydraulic support, which will not be repeated here.

[0139] The present invention firstly arranges the magnetostrictive displacement sensor in the column cylinder 3 based on the characteristics of non-contact continuous measurement, long service life, high precision, high repeatability measurement, anti-radio frequency interference, and no need for regular calibration and maintenance. The interference problem between the displacement sensors is avoided by the designed installation position. Based on the installed displacement sensor, its temperature sensor, inclination sensor and inertial measurement unit are arranged in the electronic compartment 14 of the displacement sensor to collect status data; then, based on the conditions of the working environment of the hydraulic support, environmental data is collected to correct and compensate for the key parameters for calculating the height of the hydraulic support: the inclination angle of the column cylinder 3 and the displacement of the piston in the column cylinder 3, so as to eliminate environmental interference and reduce measurement errors.

[0140] The present invention includes the following steps in the correction process of the inclination angle of the column cylinder 3: based on the two inclination sensors set up in the structure, the weights of the two sensors in the angle value calculation are allocated through the temperature influence coefficient of the sensor itself, so as to eliminate temperature interference and sensor error; based on the inclination correction value and the attitude angle data of the column cylinder 3 measured by the inertial measurement unit, data fusion is performed to comprehensively utilize the advantages of both and improve the overall accuracy and precision of the attitude angle measurement. The process of displacement compensation includes: based on the main external influencing factors of the magnetostrictive displacement sensor error, the temperature data and vibration data are measured, and the vibration data is decomposed through the inclination data finally obtained to obtain the direct parameters of the forced vibration inside the sensor, and finally the measured data and the data of the sensor itself are used as input to construct an error compensation model, obtain displacement compensation data, and reduce the influence of environmental interference in the application scenario on the sensor accuracy.

[0141] The above description is only a preferred embodiment of the present invention and does not limit the present invention in any form. Although the present invention has been disclosed as a preferred embodiment as above, it is not used to limit the present invention. Any technical personnel in this field can make some changes or modify the technical contents disclosed above into equivalent embodiments without departing from the scope of the technical solution of the present invention. However, any simple modification, equivalent change and modification made to the above embodiments according to the technical essence of the present invention without departing from the content of the technical solution of the present invention still fall within the scope of the technical solution of the present invention.

Claims

1. A method for measuring the height of a hydraulic support based on magnetostrictive displacement measurement, characterized in that: The following steps are involved: Obtaining status data: collecting inclination data, temperature data and vibration data through the inclination sensor, temperature sensor and inertial measurement unit set in the electronic compartment of the magnetostrictive displacement sensor, and sending the data to the data processing unit through the Bluetooth signal transceiver; Data preprocessing: The data processing unit preprocesses the received status data, performs temperature-based weighted averaging on the inclination data and fuses it with the attitude angle data obtained by the inertial measurement unit using the Kalman filter algorithm. The vibration data is decomposed through the inclination fusion value obtained by fusion to obtain the vibration value perpendicular to the inclination direction; Obtaining displacement data: Taking temperature data, vibration data, displacement data before compensation and the resonant frequency of the waveguide as input and displacement data after compensation as output, an error compensation model is established, wherein the displacement data before compensation is obtained by a magnetostrictive displacement sensor; The displacement data obtained by the magnetostrictive displacement sensor is compensated by the error compensation model based on the collected temperature data and vibration data decomposition values ​​to obtain the displacement data of the column cylinder piston; Calculate the support height: Use the obtained inclination fusion value and displacement compensation value to calculate the height of the hydraulic support using trigonometric functions.

2. The method for measuring the height of a hydraulic support based on magnetostrictive displacement measurement according to claim 1, characterized in that: In the step of calculating the support height, the height calculation formula of the hydraulic support is: ; In the formula, H is the height of the hydraulic support, L is the total length of the column cylinder after displacement, A It is the inclination fusion value of the column cylinder.

3. The method for measuring the height of a hydraulic support based on magnetostrictive displacement measurement according to claim 1 is characterized in that: Before the step of obtaining status data, the following steps are also included: Installation of displacement sensor: According to the structure of the hydraulic support column cylinder, the first magnetostrictive displacement sensor is coaxially installed inside the cylinder barrel to measure the displacement of the first-stage piston; the second magnetostrictive displacement sensor is installed on one side of the center axis of the second-stage piston rod to measure the displacement of the second-stage piston.

4. The method for measuring the height of a hydraulic support based on magnetostrictive displacement measurement according to claim 3 is characterized in that: The column oil cylinder comprises a cylinder barrel, a primary piston and a primary piston rod are movably arranged in the cylinder barrel, a secondary piston and a secondary piston rod are movably arranged in the primary piston rod; the primary piston and the primary piston rod are integrally connected, and the secondary piston and the secondary piston rod are integrally connected; The first magnetostrictive displacement sensor and the second magnetostrictive displacement sensor both include a waveguide wire, a measuring rod, an electronic compartment and a magnetic ring; each of the electronic compartments is provided with an inclination sensor, a temperature sensor and an inertial measurement unit; The electronic compartment of the first magnetostrictive displacement sensor is embedded in the cylinder barrel, the measuring rod of the first magnetostrictive displacement sensor passes through the primary piston and extends into the secondary piston rod, and the magnetic ring 1 of the first magnetostrictive displacement sensor is embedded in the end surface center of the primary piston; the electronic compartment of the second magnetostrictive displacement sensor is embedded in the end of the secondary piston rod, the measuring rod of the second magnetostrictive displacement sensor is installed on one side of the secondary piston rod and extends inward, and the magnetic ring 2 of the second magnetostrictive displacement sensor is embedded in the end surface center of the primary piston rod and close to one side of the measuring rod of the second magnetostrictive displacement sensor; the magnetic ring 1 of the first magnetostrictive displacement sensor is annular, and the magnetic ring 2 of the second magnetostrictive displacement sensor is an open semi-annular.

5. The method for measuring the height of a hydraulic support based on magnetostrictive displacement measurement according to claim 4 is characterized in that: The inclination sensor is used to measure the inclination angle of the column cylinder; the temperature sensor is used to measure the temperature of the first magnetostrictive displacement sensor and the second magnetostrictive displacement sensor; the inertial measurement unit is used to measure the posture angle data of the column cylinder and the intensity and direction of mechanical vibration.

6. The method for measuring the height of a hydraulic support based on magnetostrictive displacement measurement according to claim 3 is characterized in that: The data preprocessing step includes the following sub-steps: Calculating inclination data: calculating the inclination correction value of the column cylinder according to the temperature compensation model through the temperature data and inclination data respectively obtained by the first magnetostrictive displacement sensor and the second magnetostrictive displacement sensor; Attitude angle fusion: The attitude angle data of the column cylinder is measured in real time through the inertial measurement unit, and the attitude angle data is filtered through the Kalman filter algorithm. The state of a system is estimated by underestimation, and the inclination angle fusion value is obtained by fusing the attitude angle data and the inclination correction value; Obtain vibration data: Obtain vibration data through an inertial measurement unit, decompose the vibration data according to the inclination fusion value, and obtain the vibration value perpendicular to the inclination direction, where the vibration value includes vibration intensity and vibration direction.

7. The method for measuring the height of a hydraulic support based on magnetostrictive displacement measurement according to claim 6, characterized in that: The temperature compensation model is expressed as: ; In the formula, is the inclination correction value, and are the inclination values ​​measured by the inclination sensors in the first magnetostrictive displacement sensor and the second magnetostrictive displacement sensor, respectively. and are the temperature values ​​measured by the temperature sensors of the first magnetostrictive displacement sensor and the second magnetostrictive displacement sensor, respectively, and are reference temperatures of the inclination sensors corresponding to the first magnetostrictive displacement sensor and the second magnetostrictive displacement sensor respectively; and are the coefficients of the polynomial, N is the order of the polynomial, and is the weight coefficient, which is allocated according to the temperature influence coefficient of the inclination sensor.

8. The method for measuring the height of a hydraulic support based on magnetostrictive displacement measurement according to claim 6, characterized in that: The posture angle fusion step includes the following sub-steps: System state model: define the state vector M , including the state variables of the inclination angle, the inclination angle and the angular velocity, expressed as: ; System dynamics model: Under the assumption of a linear system, using the state transfer matrix A and the input matrix B Represents the system's state update process, describing the process of how the system state evolves over time m k : m k = A * m k-1 + B * u k + w k ; In the formula, u k For external input, w k is the process noise; Observation Model: Using the Observation Matrix H and observation noise v k Characterization Measurement Values z k , define an observation equation: z k = H * m k + v k ; Kalman filter: At every moment k , using the system dynamics model to predict the estimated value and covariance matrix of the current state; Combining the observation model and sensor measurements, the predicted value is corrected by the Kalman gain to obtain the final state estimate and covariance matrix; By comparing the measured data with the predicted values, the Kalman gain calculation adjusts the predicted values.

9. The method for measuring the height of a hydraulic support based on magnetostrictive displacement measurement according to claim 1, characterized in that: The step of obtaining displacement data comprises the following sub-steps: Establishing a data set: Apply vibration interference perpendicular to the measuring rod under different temperature environments, measure the measurement data of the magnetostrictive displacement sensor, and calibrate it as label data through high-precision measuring equipment to establish a data set; Data processing: clean the data set, remove or correct missing values ​​and outliers, extract relevant features from the original data, and normalize the features; Construct an error compensation model: take temperature data, vibration data, displacement data before compensation and the resonant frequency of the waveguide as input, and displacement data after compensation as output to establish an error compensation model; Model training: Input the optimized hyperparameters into the error compensation model, use the data set to train and evaluate the model, determine the fitness of the current hyperparameters, and adjust the model parameters to minimize the error; Displacement data compensation: The collected and processed real-time temperature data, vibration data, displacement data before compensation, and the waveguide wire resonant frequency of the magnetostrictive displacement sensor are input into the error compensation model to obtain the compensated displacement data.

10. The method for measuring the height of a hydraulic support based on magnetostrictive displacement measurement according to claim 9, characterized in that: The step of constructing an error compensation model includes the following sub-steps: The least squares support vector machine model is established, and the regression function is: ; In the formula, l is the number of samples, xi For the i The input of the iteration, x is the input data set, b is the bias term, αi is the Lagrange multiplier, k ( x , xi ) is the kernel function that satisfies the Mercer condition; Select the kernel function, where the kernel function K The expression is: K = aK poly +(1- a ) K RBF ; In the formula, K poly is the polynomial kernel function, K RBF is the RBF kernel function, a is the mixed weight coefficient of the polynomial kernel function and the RBF kernel function; The RBF kernel function parameters in the hybrid function are optimized by particle swarm optimization algorithm. γ Value, penalty coefficient c Values, polynomial kernel function parameters q Values ​​and the mixed weight coefficients of the two kernel functions a Optimize Set the particle swarm size and randomly generate a group of ( γ , c , q , a ) as the initial position of the particle; determine the value range of the parameter to be optimized and set the maximum speed; The fitness value of each particle in the particle swarm F ( x i ) is used for calculation, the particle swarm is divided into several subgroups, denoted as S, the optimal particle with the best fitness value in the S subgroups is calculated, the optimal solution of each subgroup in the S subgroups is obtained, and the particle with the best fitness value in the subgroup is selected to form a population gene library; Among them, the expression of the fitness value function is: ; In the formula, yi For the i The measured value of samples; fi For the i The predicted value of samples, m is the total size of the particle swarm; Calculate the shrinkage and expansion coefficients and the learning probability parameters of the subgroup, and iteratively update the fitness value of the particles and the optimal solution of the population; When it iterates to the evolutionary cycle, the inferior particles in the sub-group are eliminated and the gene pool of the established population is updated; Return to the shrinkage and expansion coefficients and the learning probability parameter steps of the subgroup, repeat the above steps until the iteration is completed, solve the optimal solution of the population, and obtain the optimal value of the hyperparameter.