Intelligent support system and intelligent support system monitoring method

Abstract

The invention relates to the technical field of bridge monitoring, and provides an intelligent support system and an intelligent support system monitoring method.The intelligent support system comprises a support and an electrical bin, a plurality of vertical displacement sensors, a plurality of horizontal displacement sensors and a plurality of temperature sensors are arranged on the support, the support comprises a steel bottom basin, and the steel bottom basin is connected with a sliding plate; a plurality of first pressure sensors are arranged at the bottom of the pressure-bearing pad; a plurality of second pressure sensors are arranged on the side wall of the pressure-bearing pad; the top of the pressure-bearing pad is connected with a middle lining plate; the top of the middle lining plate is connected with a spherical crown lining plate; and a sensor acquisition module, a communication module, a double-shaft tilt angle sensor, a data processing module and an in-situ calibration module are arranged in the electrical cabin. By adopting the structure, multi-dimensional data can be monitored, and the data accuracy and long-term reliability are improved.

Description

Technical Field

[0001] This invention relates to the field of bridge monitoring technology, and in particular to an intelligent bearing system and a monitoring method for the intelligent bearing system. Background Technology

[0002] As a key force-transmitting component connecting the superstructure and substructure of a bridge, the working condition of bridge bearings directly affects the safety and durability of the entire bridge structure.

[0003] In practical engineering, bridge bearings are subject to various complex external forces, including static loads, dynamic loads, temperature stress, and seismic loads, over long periods of time. This makes them prone to stress concentration, aging failure, and excessive displacement. Existing bridge bearings typically integrate sensors to monitor their physical parameters in real time, thus preventing bridge structural safety accidents.

[0004] Existing intelligent bearing systems have relatively limited measurement dimensions for physical parameters. For example, Chinese patent 201621075918.3 only involves a combination of pressure and displacement sensors, lacking the ability to monitor the key parameter of bearing rotation angle. Furthermore, the sensor layout is limited to the side edge of the bearing, resulting in low sensor integration and an inability to obtain the stress distribution inside the bearing. Although Chinese patent 202410135926.5 adds monitoring of rotation angle, it lacks a means to intelligently calibrate the sensor data. During long-term service, factors such as bearing temperature and sensor errors can easily lead to inaccurate monitoring data. Moreover, sensor calibration requires returning the system to the factory. With multiple sensors used in the system, the maintenance time is long, and long-term reliability is lacking. Summary of the Invention

[0005] To address the shortcomings of existing technologies, the present invention aims to provide an intelligent bearing system and a monitoring method for such a system. This invention utilizes a vertical displacement sensor, a horizontal displacement sensor, a temperature sensor, a dual-axis tilt sensor, a pressure sensor, and an in-situ calibration module to achieve long-term, reliable, and accurate data monitoring. Multiple curved surface structures improve the uniformity of stress distribution during fabrication, enhancing the reliability of the bearing system. Furthermore, the monitoring results are evaluated using various data sources. This invention aims to solve the technical problems of existing intelligent bearing systems, such as limited data dimensions, insufficient data accuracy, inconvenient maintenance, and insufficient long-term reliability.

[0006] To achieve the above objectives, the present invention is implemented through the following technical solution: A smart support system includes a support and an electrical compartment connected to the support. The support includes a steel base, an intermediate liner, and a spherical crown liner. The top surface of the steel base is a first arc surface that is concave to the bottom. A sliding plate is slidably connected to the first arc surface. The bottom surface of the sliding plate is adapted to the first arc surface. A pressure pad is connected to the top of the sliding plate. The top of the pressure pad is connected to the intermediate liner. The top of the intermediate liner is connected to the spherical crown liner. The top surface of the intermediate liner is a second arc surface that is concave to the bottom. The bottom of the spherical crown liner convexes outward to form a third arc surface. The third arc surface has the same curvature as the second arc surface. A sensor group is installed on the support. The sensor group is used to collect initial displacement data, initial temperature data, and... Initial pressure data is collected within the electrical compartment, which is equipped with a sensor acquisition module, a dual-axis tilt sensor, a data processing module, and an in-situ calibration module. The dual-axis tilt sensor is used to collect the initial rotation angle data of the support. The sensor acquisition module is connected to the sensor group and the dual-axis tilt sensor to obtain an initial sensor dataset. The data processing module is connected to the sensor acquisition module and the in-situ calibration module. The data processing module processes the initial sensor dataset to obtain the support eccentricity, actual rotation center, and performance degradation rate, and constructs a membership cloud model. The support eccentricity, actual rotation center, and performance degradation rate are input into the membership cloud model to obtain the monitoring and evaluation results of the support.

[0007] Furthermore, the sensor group includes several vertical displacement sensors, several horizontal displacement sensors, several temperature sensors, several first pressure sensors, and several second pressure sensors. Several first pressure sensors are disposed at the bottom of the pressure pad, and several second pressure sensors are disposed on the sidewall of the pressure pad. The first pressure sensors are used to collect vertical pressure data of the support, and the second pressure sensors are used to collect lateral deformation pressure data of the pressure pad. The in-situ calibration module includes a force signal generator and a displacement generator. The displacement generator is connected to several vertical displacement sensors and several horizontal displacement sensors, and the force signal generator is connected to several first pressure sensors and several second pressure sensors.

[0008] Furthermore, a circular groove is formed on the side of the sliding plate facing away from the steel base, the bottom of the circular groove is connected to the pressure pad, a sealing ring is fitted on the outer side wall of the sliding plate, the bottom of the sealing ring is adapted to the first arc surface, the support also includes an upper support plate, the top of the spherical crown liner is connected to the upper support plate, and the side of the upper support plate is connected to the side of the steel base through several connectors.

[0009] Furthermore, the skateboard is made of polytetrafluoroethylene, the pressure pad is made of rubber, and the vertical displacement sensors are arranged in a ring array.

[0010] Furthermore, the communication module includes a communication board, on which LoRa, 4G and Beidou communication units are connected. A power board is installed inside the electrical compartment, which is connected to a solar cell and a supercapacitor charging and discharging unit. The sensor acquisition module is connected to several vertical displacement sensors, several horizontal displacement sensors, several temperature sensors, several first pressure sensors and several second pressure sensors via aviation connectors. A debugging interface is provided on the electrical compartment.

[0011] A monitoring method for an intelligent bearing system, using the intelligent bearing system described in the above technical solution, the method comprising the following steps: The sensor array was calibrated using an in-situ calibration module. The sensor group collects the initial displacement data, initial temperature data, and initial pressure data of the support, and the dual-axis tilt sensor collects the initial rotation angle data of the support, so as to form an initial sensor dataset by combining the initial displacement data, the initial temperature data, the initial pressure data, and the initial rotation angle data. The initial sensor dataset is acquired by the sensor acquisition module and transmitted to the data processing module; The initial sensor dataset is processed by the data processing module to obtain the support eccentricity, actual rotation center, and performance degradation rate. The membership cloud model is constructed through the data processing module. The eccentricity of the support, the actual rotation center, and the performance degradation rate are input into the membership cloud model to obtain the monitoring and evaluation results of the support, so as to provide remote early warning based on the monitoring and evaluation results.

[0012] Furthermore, the sensor group includes several vertical displacement sensors, several horizontal displacement sensors, several first pressure sensors, and several second pressure sensors. The in-situ calibration module includes a force signal generator and a displacement generator. The step of calibrating the system parameters of the sensor group using the in-situ calibration module includes: The force signal generator in the in-situ calibration module is used to apply several standard force signals to several first pressure sensors and several second pressure sensors respectively to obtain several pressure response data. Several pressure response data are compared with the standard force signal, and several pressure calibration coefficients are calculated by the least squares method. The system parameters of several first pressure sensors and several second pressure sensors are adjusted according to the several pressure calibration coefficients. The displacement generator in the in-situ calibration module is used to apply several standard displacements to several vertical displacement sensors and several horizontal displacement sensors respectively to obtain several displacement response data. Several displacement response data are compared with the standard displacement, and several displacement calibration coefficients are calculated using the least squares method. The system parameters of several vertical displacement sensors and several horizontal displacement sensors are adjusted according to the several displacement calibration coefficients.

[0013] Furthermore, the step of processing the initial sensor dataset through the data processing module to obtain the support eccentricity, actual rotation center, and performance degradation rate includes: The data processing module performs moving average filtering and wavelet denoising on the initial sensor dataset and removes outliers to obtain a monitoring sensor dataset, which includes monitoring temperature data, monitoring pressure data, monitoring displacement data and monitoring rotation angle data. Temperature compensation calculations are performed on the monitored pressure data, monitored displacement data, and monitored rotation angle data based on the monitored temperature data to obtain compensated pressure data, compensated displacement data, and compensated rotation angle data; Based on the compensation pressure data, the support eccentricity is extracted; based on the compensation displacement data and the compensation rotation angle data, the actual rotation center is extracted; historical monitoring data is obtained; and based on the historical monitoring data, the compensation pressure data, the compensation displacement data, and the compensation rotation angle data, the performance degradation rate is obtained.

[0014] Furthermore, the step of constructing a membership cloud model through the data processing module, and inputting the support eccentricity, the actual rotation center, and the performance degradation rate into the membership cloud model to obtain the monitoring and evaluation results of the support includes: The data processing module obtains the training dataset and hyperentropy coefficients. Based on the training dataset and hyperentropy coefficients, the mean method is used to determine the cloud model expectation and cloud model hyperentropy in order to construct a cloud generator. The cloud generator includes a cloud membership function, which is used to calculate the cloud membership degree. A fuzzy neural network is obtained, and the cloud membership degree is used as the input of the fuzzy neural network to form a membership degree cloud model by combining the cloud generator and the fuzzy neural network. The bearing eccentricity, the actual rotation center, and the performance degradation rate are input into the membership cloud model to obtain the monitoring and evaluation results of the bearing.

[0015] Furthermore, the cloud membership function is:

[0016] in, Indicates the membership degree of the cloud. Indicates the input value. Indicates the number of cloud droplets;

[0017] in, Represents the normal random membership degree. This represents the expected value of the cloud model. This indicates that the cloud model expectation and cloud model hyperentropy are used as the expectation and variance, respectively, to calculate the first... A normal random number.

[0018] Compared with the prior art, the beneficial effects of the present invention are as follows: Through the mutual matching of the arc surfaces between the steel base, the sliding plate, the intermediate liner, and the spherical crown liner, the force on the support is made more uniform, and self-centering adjustment can be achieved during installation; by setting the vertical displacement sensor, the horizontal displacement sensor, the temperature sensor, the first pressure sensor, the second pressure sensor, and the dual-axis tilt sensor, multi-dimensional data of the support is collected. The first and second pressure sensors are set at different positions on the bearing pad, which can monitor not only vertical pressure but also lateral deformation pressure on the bearing pad, providing more comprehensive monitoring of the bridge support; by setting the in-situ calibration module, the pressure sensors and displacement sensors are calibrated. The sensor achieves in-situ calibration, eliminating the need to return a large number of sensors in the system to the factory for calibration, thus ensuring the long-term reliability of the intelligent support system and significantly reducing the manpower, material resources, and time spent on sensor maintenance. Through the data processing module, temperature compensation calculation is performed on the sensor monitoring data, making the monitoring data more accurate. It also enables intelligent analysis and evaluation of multi-dimensional data such as pressure, displacement, rotation angle, and temperature in the intelligent support system. Specifically, the membership cloud model generates random changes when calculating the cloud membership degree, improving the hardened boundary of the traditional membership function. This further satisfies the randomness and fuzziness of qualitative and quantitative transformation in fuzzy inference systems, making it more consistent with the characteristics of fuzzy inference, thereby obtaining an evaluation result that is closer to the actual support condition. Attached Figure Description

[0019] Figure 1 This is a schematic diagram of the structure of the support in the intelligent support system according to the first embodiment of the present invention; Figure 2 This is a schematic diagram of the electrical compartment in the intelligent support system according to the first embodiment of the present invention; Figure 3 This is a partial disassembly diagram of the intelligent support system in the first embodiment of the present invention; Figure 4This is a schematic diagram of the steel base, sliding plate, and intermediate liner in the intelligent support system of the first embodiment of the present invention; Figure 5 This is a schematic diagram of the steel base and sliding plate in the intelligent support system according to the first embodiment of the present invention; Figure 6 This is a schematic diagram of the structure of the spherical crown liner and intermediate liner in the intelligent support system according to the first embodiment of the present invention; Figure 7 This is a flowchart of the intelligent support monitoring method in the second embodiment of the present invention; Explanation of key component symbols: 100. Steel base plate; 110. Slide plate; 111. Circular groove; 112. Pressure pad; 120. Sealing ring; 130. Intermediate liner plate; 140. Spherical crown liner plate; 150. Upper support plate; 160. Connector; 200. Electrical compartment; 210. Sensor interface; 220. Debugging interface.

[0020] The following detailed description, in conjunction with the accompanying drawings, will further illustrate the present invention. Detailed Implementation

[0021] To facilitate understanding of the present invention, a more complete description will be given below with reference to the accompanying drawings. Several embodiments of the invention are illustrated in the drawings. However, the invention can be implemented in many different forms and is not limited to the embodiments described herein. Rather, these embodiments are provided so that this disclosure will be thorough and complete.

[0022] It should be noted that when a component is said to be "fixed to" another component, it can be directly on the other component or there may be an intervening component. When a component is said to be "connected to" another component, it can be directly connected to the other component or there may be an intervening component. The terms "vertical," "horizontal," "left," "right," and similar expressions used in this document are for illustrative purposes only.

[0023] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention pertains. The terminology used herein in the description of the invention is for the purpose of describing particular embodiments only and is not intended to be limiting of the invention. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0024] Please see Figures 1 to 6The intelligent support system in the first embodiment of the present invention includes a support and an electrical compartment 200 connected to the support. A sensor group is installed on the support, which is used to collect initial displacement data, initial temperature data, and initial pressure data of the support. The sensor group includes several vertical displacement sensors, several horizontal displacement sensors, several temperature sensors, several first pressure sensors, and several second pressure sensors. The support includes a steel base 100, the top surface of which is a first arc surface concave towards the bottom. A sliding plate 110 is slidably connected to the first arc surface, and the bottom surface of the sliding plate 110 is adapted to the first arc surface. The outer wall of the sliding plate 110 is fitted with... A sealing ring 120 is connected, the bottom of which is adapted to the first arc surface. A circular groove 111 is formed on the side of the sliding plate 110 facing away from the steel base 100. The bottom of the circular groove 111 is connected to a pressure pad 112. Several first pressure sensors are arranged on the bottom of the pressure pad 112, and several second pressure sensors are arranged on the side wall of the pressure pad 112. The sliding plate 110 is made of polytetrafluoroethylene, and the pressure pad 112 is made of rubber. The first pressure sensors are used to measure the vertical pressure of the support, and the second pressure sensors are used to measure the lateral deformation pressure of the pressure pad 112. Several vertical displacement sensors are arranged in a ring array.

[0025] Preferably, the tilt and uneven settlement of the support can be analyzed by synchronous readings from multiple vertical displacement sensors. The horizontal displacement sensor is a wire-type distance sensor. Three temperature sensors are used to monitor the temperature changes of the support in real time. The steel base 100 includes a basin body and a lower support plate. The top surface of the lower support plate connects to the basin body. The top surface of the basin body is a concave arc surface, and the bottom surface of the sealing ring 120 is a convex arc surface, adapted to the arc surface of the basin body. Understandably, the mutual matching of the arc surfaces among the steel base 100, the sliding plate 110, the intermediate liner 130, and the spherical crown liner 140 makes the overall stress on the support more uniform, and allows for self-centering adjustment during installation, enabling convenient modular installation.

[0026] The top of the pressure pad 112 is connected to the intermediate liner 130, and the top of the intermediate liner 130 is connected to the spherical crown liner 140. The top surface of the intermediate liner 130 is a second arc surface that is concave to the bottom, and the bottom of the spherical crown liner 140 is convex to form a third arc surface. The third arc surface has the same curvature as the second arc surface. The top of the spherical crown liner 140 is connected to the upper support plate 150, and the side of the upper support plate 150 is connected to the side of the steel base basin 100 through several connectors 160. Preferably, a plurality of vertical displacement sensors are disposed on the sidewall of the upper support plate 150, forming an array around it, and a plurality of horizontal displacement sensors are disposed on the upper support plate 150 and the intermediate liner plate 130. The upper support plate 150 is a cover, its side extending downward to shield and protect part of the spherical crown liner plate 140 and the intermediate liner plate 130. A plurality of vertical displacement sensors are disposed on the sidewall of the upper support plate 150. The connector 160 includes bolts and metal rods, and the number of connectors 160 is four. It can be understood that by setting the vertical displacement sensors, the horizontal displacement sensors, the temperature sensor, the first pressure sensor, the second pressure sensor, and the dual-axis tilt sensor, multi-dimensional data of the support is collected, making the monitoring of the bridge support more comprehensive.

[0027] The electrical compartment 200 is equipped with a sensor acquisition module, a communication module, a dual-axis tilt sensor, a data processing module, and an in-situ calibration module. The dual-axis tilt sensor is used to acquire the initial rotation angle data of the support. The in-situ calibration module includes a force signal generator and a displacement generator. The displacement generator is connected to several vertical displacement sensors and several horizontal displacement sensors. The force signal generator is connected to several first pressure sensors and several second pressure sensors. The sensor acquisition module is connected to several vertical displacement sensors, several horizontal displacement sensors, several temperature sensors, several first pressure sensors, several second pressure sensors, and the dual-axis tilt sensor to obtain initial sensor data. The dataset includes a data processing module connected to the sensor acquisition module and the in-situ calibration module. The data processing module processes the initial sensor dataset and constructs a membership cloud model to output the monitoring and evaluation results of the support. The communication module includes a communication board connected to a LoRa unit, a 4G unit, and a Beidou communication unit. A power board is installed inside the electrical compartment 200, connected to a solar cell and a supercapacitor charging and discharging unit. The sensor acquisition module is connected to several vertical displacement sensors, several horizontal displacement sensors, several temperature sensors, several first pressure sensors, and several second pressure sensors via aviation connectors. A debugging interface 220 is provided on the electrical compartment 200.

[0028] Preferably, the electrical compartment 200 adopts a multi-layer shielded and sealed structure with an IP68 protection rating. The electrical compartment 200 is also equipped with a sensor interface 210, which connects to an aviation connector. The electrical compartment 200 internally employs a layered board design, housing a main control board based on an ARM Cortex-M7 core with a 600MHz clock speed, and equipped with 1MB SRAM and 2MB RAM. The FLASH sensor acquisition module includes a sensor acquisition board containing a 16-channel 24-bit high-precision ADC with a maximum sampling rate of 1kHz. The LoRa unit supports the 470MHz frequency band. The supercapacitor charging and discharging unit incorporates a supercapacitor, which combines the rapid charging and discharging characteristics of a capacitor with the energy storage characteristics of a battery. The power board supports solar cell input and supercapacitor charging and discharging management through the solar cell and the supercapacitor charging and discharging unit. The debugging interface 220 is used for on-site data export and parameter configuration. By setting up the in-situ calibration module, the pressure sensor and displacement sensor can be calibrated in situ, eliminating the need to return a large number of sensors in the system to the factory for calibration, ensuring the long-term reliability of the intelligent support system, and significantly reducing the manpower, material resources, and time spent on sensor maintenance. Through the data processing module, the membership cloud model is constructed, combining the cloud generator and neural network to achieve intelligent analysis and evaluation of multi-dimensional data such as pressure, displacement, rotation angle, and temperature in the intelligent support system.

[0029] Furthermore, membership degree refers to the degree to which an element belongs to or does not belong to a set. It can be used to quantitatively describe fuzzy concepts. Existing intelligent analysis and evaluation methods often use membership functions, which are key to quantifying qualitative concepts. Membership functions evaluate and rate input data, avoiding the subjectivity of manual evaluation. However, traditional membership functions have the problem of overly rigid boundaries. At the same time, it is difficult to determine a membership function with good applicability in intelligent evaluation. Therefore, the use of traditional membership functions lacks randomness and fuzziness. When combined with neural networks to form a fuzzy inference system, its overly rigid boundary characteristics cannot meet the requirements of randomness and fuzziness in fuzzy inference. Understandably, by constructing the membership cloud model, the traditional membership function is softened. Through normally distributed random numbers, the objectivity of fuzzy concepts is effectively enhanced, meeting the fuzziness requirements of fuzzy neural networks, which is beneficial to obtaining intelligent analysis and evaluation results that are more consistent with the real situation.

[0030] Furthermore, a pressure test was conducted on the intelligent bearing system described in this embodiment based on a 1000kN range. The pressure measurement accuracy of a traditional intelligent bearing system is ±2.5%FS, while that of the intelligent bearing system in this embodiment is ±2.5%FS. Based on a displacement test with a 50mm range, the pressure measurement accuracy of a traditional intelligent bearing system is ±1.5mm, while that of the intelligent bearing system in this embodiment is ±0.3mm. The rotation angle measurement accuracy of the intelligent bearing system in this embodiment is ±0.05°. Based on a temperature test in an environment of -20℃ to 60℃, the pressure monitoring accuracy of a traditional intelligent bearing system is affected by temperature by 0.05% FS / ℃, while that of the intelligent bearing system in this embodiment is affected by temperature by 0.01% FS / ℃. Based on an accelerated aging test, a traditional intelligent bearing system needs to be replaced every 3 years, while the intelligent bearing system in this embodiment can achieve 10 years of maintenance-free operation. Understandably, the intelligent support system significantly improves measurement accuracy, measurement precision, service life, and long-term reliability compared to traditional intelligent support systems, while also enhancing the ease of maintenance.

[0031] Please see Figure 7 The intelligent bearing system monitoring method in the second embodiment of the present invention is applied to the intelligent bearing system as described in the first embodiment, and the method includes the following steps: Step S10: Use the in-situ calibration module to calibrate the system parameters of the sensor group; Understandably, by performing in-situ calibration on multiple sensors in the system, errors in the sensors during long-term service are avoided, and the need for sensors to be returned to the factory for calibration in traditional technical solutions is avoided. This significantly improves the accuracy and long-term reliability of sensor data acquisition, and saves manpower, material resources and maintenance time.

[0032] The sensor group includes several vertical displacement sensors, several horizontal displacement sensors, several first pressure sensors, and several second pressure sensors. The in-situ calibration module includes a force signal generator and a displacement generator. Step S10 includes: S110: The force signal generator in the in-situ calibration module is used to apply several standard force signals to several first pressure sensors and several second pressure sensors respectively to obtain several pressure response data. S120: Compare several pressure response data with the standard force signal respectively, calculate several pressure calibration coefficients by the least squares method, and adjust the system parameters of several first pressure sensors and several second pressure sensors according to several pressure calibration coefficients. S130: The displacement generator in the in-situ calibration module is used to apply several standard displacements to several vertical displacement sensors and several horizontal displacement sensors respectively to obtain several displacement response data. S140: Compare the several displacement response data with the standard displacement respectively, calculate several displacement calibration coefficients by the least squares method, and adjust the system parameters of several vertical displacement sensors and several horizontal displacement sensors according to the several displacement calibration coefficients.

[0033] Preferably, the first pressure sensor, the second pressure sensor, the vertical displacement sensor, and the horizontal displacement sensor all use a least squares straight line as the working straight line of the sensor. Several calibration points are set within the sensor's measurement range, and several cyclic calibrations are performed to obtain several measurement values. The working straight line is obtained by combining the least squares straight line equation. The slope of the working straight line represents the sensor's sensitivity. The sensor's full-scale output value is calculated by the slope of the working straight line, the upper limit of the measurement range, and the lower limit of the measurement range. Based on the difference between the response data and the standard data, the differences in parameters such as sensitivity, full-scale output value, and zero point between the current state sensor and the standard state sensor can be obtained through the least squares straight line equation, thereby obtaining the calibration coefficient and adjusting the sensor's system parameters. The intelligent support system can be put into calibration mode by remote command or on-site triggering. The system records the current data as a reference. The several standard force signals are 50kN, 100kN, and 150kN signals, and the several standard displacements are 1mm, 2mm, and 3mm, respectively.

[0034] Preferably, after the steps of comparing the plurality of displacement response data with the standard displacement, calculating the plurality of displacement calibration coefficients using the least squares method, and adjusting the system parameters of the plurality of vertical displacement sensors and the plurality of horizontal displacement sensors according to the plurality of displacement calibration coefficients, the method further includes: applying the plurality of verification force signals to the plurality of first pressure sensors and the plurality of second pressure sensors to obtain the plurality of pressure verification data; applying the plurality of verification displacements to the plurality of vertical displacement sensors and the plurality of horizontal displacement sensors to obtain the plurality of displacement verification data; obtaining the pressure verification measurement accuracy based on the plurality of verification force signals and the plurality of pressure verification data; obtaining the displacement verification measurement accuracy based on the plurality of verification displacements and the plurality of displacement verification data; obtaining the pressure measurement accuracy threshold and the displacement measurement accuracy threshold; comparing the pressure measurement accuracy threshold and the pressure verification measurement accuracy; comparing the displacement measurement accuracy threshold and the displacement verification measurement accuracy to determine whether the calibration is complete. Furthermore, two verification force signals, namely a 75kN signal and a 125kN signal, are used, and two verification displacements, namely a 1.5mm displacement and a 2.5mm displacement, are used. The pressure measurement accuracy threshold is 1%, and the displacement measurement accuracy threshold is 2%.

[0035] Step S20: Collect initial displacement data, initial temperature data, and initial pressure data of the support through the sensor group, and collect initial rotation angle data of the support through the dual-axis tilt sensor, so as to form an initial sensor dataset by combining the initial displacement data, the initial temperature data, the initial pressure data, and the initial rotation angle data. Understandably, this enables the system to collect multi-dimensional data.

[0036] Step S30: Obtain the initial sensor dataset through the sensor acquisition module and transmit it to the data processing module; Step S40: The initial sensor dataset is processed by the data processing module to obtain the support eccentricity, actual rotation center, and performance degradation rate; Step S40 includes: S410: The data processing module performs moving average filtering and wavelet denoising on the initial sensor dataset and removes outliers to obtain a monitoring sensor dataset, which includes monitoring temperature data, monitoring pressure data, monitoring displacement data and monitoring rotation angle data. S420: Based on the monitored temperature data, perform temperature compensation calculations on the monitored pressure data, the monitored displacement data, and the monitored rotation angle data to obtain compensated pressure data, compensated displacement data, and compensated rotation angle data; Preferably, the temperature-output characteristic curve is calibrated based on historical data, and a piecewise linear interpolation method is used to perform temperature compensation calculation on the sensor output data based on the monitored temperature data. Understandably, the sensor data will have certain errors at different temperatures. Through timely temperature compensation calculation, the collected sensor monitoring data can be made more accurate.

[0037] S430: Based on the compensation pressure data, extract the support eccentricity; based on the compensation displacement data and the compensation rotation angle data, extract the actual rotation center; obtain historical monitoring data; and based on the historical monitoring data, the compensation pressure data, the compensation displacement data, and the compensation rotation angle data, obtain the performance degradation rate.

[0038] Preferably, the first pressure sensor is used to collect vertical pressure data of the support, and the second pressure sensor is used to collect lateral deformation pressure data of the bearing pad. Several first pressure sensors are arranged in a circular array. When the lateral deformation pressure data indicates lateral deformation of the support bearing pad, the vertical pressure data is corrected using a lateral correction coefficient and the lateral deformation pressure data to eliminate local distortion. The force center is obtained by weighted averaging of the corrected vertical pressure data, thereby calculating the eccentricity between the force center and the geometric center of the support, i.e., the support eccentricity. Several vertical displacement sensors are installed in a circular array on the sidewall of the upper support plate. The location of the dual-axis tilt sensor is taken as the geometric center of the support. The compensated rotation angle data includes a first rotation angle and a second rotation angle. In the case of a small rotation angle, the displacement of the vertical displacement sensor is caused by the rotation of the actual rotation center. The compensated displacement data represents the actual position relative to the bearing pad. The difference in the initial installation position, based on the vertical compensation displacement data of the vertical displacement sensor, the installation coordinates, the first rotation angle, and the second rotation angle, yields the rotation center coordinates. These coordinates are then verified by combining the horizontal compensation displacement data of the horizontal displacement sensor, the first rotation angle, and the second rotation angle. The historical monitoring data includes historical support eccentricity, historical rotation center offset, historical maximum vertical displacement, and historical pressure distribution unevenness at multiple historical moments. Combining the existing compensation pressure data, compensation displacement data, and compensation rotation angle data, the eccentricity degradation, rotation center offset degradation, pressure distribution unevenness degradation, and maximum vertical displacement degradation are calculated and normalized. Specifically, the degradation amount is the difference between the current index value and the initial index value. Four weights are set using the analytic hierarchy process (AHP), and the comprehensive degradation amount is calculated by combining these four weights and the four degradation amounts. The comprehensive degradation amount divided by time gives the performance degradation rate.

[0039] Step S50: Construct a membership cloud model through the data processing module, input the support eccentricity, the actual rotation center and the performance degradation rate into the membership cloud model to obtain the monitoring and evaluation results of the support, so as to provide remote early warning based on the monitoring and evaluation results.

[0040] Preferably, the membership cloud model includes a cloud fuzzy layer, a fuzzy operator layer, a fuzzy rule layer, and an evaluation output layer. The cloud fuzzy layer is used to convert data that can characterize the support state into several membership values. The fuzzy operator layer processes different membership values ​​through fuzzy operators in the fuzzy neural network and outputs weights of multiple fuzzy rules. Fuzzy rules specifically refer to the reasoning rules followed between input and output in the fuzzy inference process. The fuzzy rule layer is used to construct a fuzzy set with multiple fuzzy rules. The rules adopt the if-then type, with each rule corresponding to an if condition and a then conclusion. Based on the weights of the fuzzy rules obtained from the previous layer, combined with the conclusion value, several actual outputs are calculated. The output layer then obtains the monitoring and evaluation results based on the several actual outputs.

[0041] Step S50 includes: S510: The training dataset and hyperentropy coefficients are obtained through the data processing module. Based on the training dataset and hyperentropy coefficients, the mean method is used to determine the cloud model expectation and cloud model hyperentropy in order to construct a cloud generator. The cloud generator includes a cloud membership function, which is used to calculate the cloud membership degree. The cloud membership function is:

[0042] in, Indicates the membership degree of the cloud. Indicates the input value. Indicates the number of cloud droplets;

[0043] in, Represents the normal random membership degree. This represents the expected value of the cloud model. This indicates that the cloud model expectation and cloud model hyperentropy are used as the expectation and variance, respectively, to calculate the first... A normal random number.

[0044] Preferably, the cloud generator is a forward cloud generator, constructed based on a normal cloud model. The mean of the maximum and minimum values ​​in the training dataset is calculated, and the mean, the expected value of the cloud model, and the super-entropy of the cloud model are used as digital features of the forward cloud generator. A number of normally distributed random numbers are generated for each input value using a normal cloud algorithm. The Gaussian membership degrees corresponding to these normally distributed random numbers are then calculated, which are the normally distributed random membership degrees. Each input value and its corresponding normally distributed random membership degree are considered as a cloud droplet, and the cloud generator includes several cloud droplets.

[0045] S520: Obtain a fuzzy neural network, and use the cloud membership degree as the input of the fuzzy neural network to form a membership degree cloud model by combining the cloud generator and the fuzzy neural network; Preferably, the fuzzy neural network is an adaptive neural-fuzzy inference network.

[0046] S530: Input the bearing eccentricity, the actual rotation center, and the performance degradation rate into the membership cloud model to obtain the monitoring and evaluation results of the bearing.

[0047] Preferably, the step of inputting the support eccentricity, the actual rotation center, and the performance degradation rate into the membership cloud model to obtain the monitoring and evaluation results of the support includes: inputting the support eccentricity, the actual rotation center, and the performance degradation rate into the cloud generator in the membership cloud model to calculate the first cloud membership degree, the second cloud membership degree, and the third cloud membership degree respectively; generating a plurality of fuzzy rule excitations based on the first cloud membership degree, the second cloud membership degree, the third cloud membership degree, and the fuzzy operator in the fuzzy neural network; calculating a plurality of rule membership degrees based on the plurality of fuzzy rule excitations and the plurality of fuzzy rules in the fuzzy neural network; and performing a weighted average operation based on the plurality of rule membership degrees to obtain the monitoring and evaluation results of the support. Further, the fuzzy operator is a "fuzzy AND" operator, the fuzzy neural network adopts an ANFIS structure, the number of fuzzy rules is 4, the fuzzy rules adopt an if-then structure, and the adaptive parameters of each node in the fuzzy neural network are used as the parameters of the conclusion part in the fuzzy rules. Understandably, by generating random variations in the membership degree cloud model when calculating multiple cloud membership degrees, the problem of overly rigid boundaries in traditional membership functions is improved. This further satisfies the randomness and fuzziness of qualitative and quantitative transformations in fuzzy inference systems, making it more consistent with the characteristics of fuzzy inference and thus obtaining evaluation results that are closer to the actual support conditions.

[0048] In the description of this specification, references to terms such as "one embodiment," "some embodiments," "example," "specific example," or "some examples," etc., indicate that a specific feature, structure, material, or characteristic described in connection with that embodiment or example is included in at least one embodiment or example of the invention. In this specification, the illustrative expressions of the above terms do not necessarily refer to the same embodiment or example. Furthermore, the specific features, structures, materials, or characteristics described may be combined in any suitable manner in one or more embodiments or examples.

[0049] The embodiments described above are merely illustrative of several implementations of the present invention, and while the descriptions are specific and detailed, they should not be construed as limiting the scope of the present invention. It should be noted that those skilled in the art can make various modifications and improvements without departing from the concept of the present invention, and these modifications and improvements all fall within the scope of protection of the present invention. Therefore, the scope of protection of this patent should be determined by the appended claims.

Claims

1. An intelligent support system, characterized by, The support includes a steel bottom basin, an intermediate lining plate, and a spherical cap lining plate, the top surface of the steel bottom basin is a first arc surface concave to the bottom, the first arc surface is connected to a sliding plate, the bottom surface of the sliding plate is matched with the first arc surface, the top of the sliding plate is connected to a pressure bearing pad, the top of the pressure bearing pad is connected to the intermediate lining plate, the top of the intermediate lining plate is connected to the spherical cap lining plate, the top surface of the intermediate lining plate is a second arc surface concave to the bottom, the bottom of the spherical cap lining plate is outwardly convex to form a third arc surface, the third arc surface has the same curvature as the second arc surface, a sensor group is arranged on the support, the sensor group is used to collect initial displacement data, initial temperature data, and initial pressure data of the support, a sensor acquisition module, a two-axis inclination sensor, a data processing module, and an in-situ calibration module are arranged in the electrical cabinet, the two-axis inclination sensor is used to collect initial rotation angle data of the support, the sensor acquisition module is connected to the sensor group and the two-axis inclination sensor to obtain an initial sensor data set, the data processing module is connected to the sensor acquisition module and the in-situ calibration module, the data processing module is used to process the initial sensor data set to obtain a support eccentricity, an actual rotation center, and a performance degradation rate, and construct a membership cloud model, input the support eccentricity, the actual rotation center, and the performance degradation rate into the membership cloud model to obtain a monitoring evaluation result of the support.

2. The intelligent support system of claim 1, wherein, The sensor group includes a plurality of vertical displacement sensors, a plurality of horizontal displacement sensors, a plurality of temperature sensors, a plurality of first pressure sensors, and a plurality of second pressure sensors, the bottom of the pressure bearing pad is provided with a plurality of first pressure sensors, the side wall of the pressure bearing pad is provided with a plurality of second pressure sensors, the first pressure sensors are used to collect vertical pressure data of the support, the second pressure sensors are used to collect lateral deformation pressure data of the pressure bearing pad, the in-situ calibration module includes a force signal generator and a displacement generator, the displacement generator is connected to a plurality of vertical displacement sensors and a plurality of horizontal displacement sensors, and the force signal generator is connected to a plurality of first pressure sensors and a plurality of second pressure sensors.

3. The intelligent support system of claim 1, wherein, A circular groove is formed on the side of the sliding plate away from the steel bottom basin, the bottom of the circular groove is connected to the pressure bearing pad, a sealing ring is sleeved on the outer side wall of the sliding plate, the bottom of the sealing ring is matched with the first arc surface, the support further includes an upper support plate, the top of the spherical cap lining plate is connected to the upper support plate, and the side edges of the upper support plate are connected to the side edges of the steel bottom basin through a plurality of connecting pieces.

4. The intelligent support system of claim 2, wherein, The material of the sliding plate is polytetrafluoroethylene, and the material of the pressure bearing pad is rubber, and a plurality of vertical displacement sensors are arranged in a ring array.

5. The intelligent support system of claim 2, wherein, The electrical bin is internally provided with a communication module connected with the data processing module, the communication module comprises a communication board, the communication board is connected with a LoRa unit, a 4G unit and a Beidou communication unit, the electrical bin is internally provided with a power supply board connected with a solar cell and a super capacitor charging and discharging unit, the sensor acquisition module is connected with a plurality of vertical displacement sensors, a plurality of horizontal displacement sensors, a plurality of temperature sensors, a plurality of first pressure sensors and a plurality of second pressure sensors through an aviation plug, and the electrical bin is provided with a debugging interface.

6. A monitoring method of the intelligent support system, applied to the intelligent support system according to any one of claims 1-5, characterized in that, The method comprises the following steps: The in-situ calibration module is used to calibrate the system parameters of the sensor group; The initial displacement data, the initial temperature data and the initial pressure data of the support are collected by the sensor group, and the initial rotation angle data of the support is collected by the dual-axis inclination sensor, so as to form an initial sensor data set comprising the initial displacement data, the initial temperature data, the initial pressure data and the initial rotation angle data; The initial sensor data set is obtained by the sensor acquisition module and transmitted to the data processing module; The data processing module processes the initial sensor data set to obtain the support eccentricity, the actual rotation center and the performance degradation rate; The data processing module constructs a membership cloud model, inputs the support eccentricity, the actual rotation center and the performance degradation rate into the membership cloud model, obtains the monitoring evaluation result of the support, and performs remote early warning according to the monitoring evaluation result.

7. The intelligent support system monitoring method of claim 6, wherein, The sensor group comprises a plurality of vertical displacement sensors, a plurality of horizontal displacement sensors, a plurality of first pressure sensors and a plurality of second pressure sensors, the in-situ calibration module comprises a force signal generator and a displacement generator, and the step of calibrating the system parameters of the sensor group by using the in-situ calibration module comprises: The force signal generator in the in-situ calibration module is used to apply a plurality of standard force signals to the plurality of first pressure sensors and the plurality of second pressure sensors respectively, so as to obtain a plurality of pressure response data; The plurality of pressure response data are compared with the standard force signals respectively, a plurality of pressure calibration coefficients are calculated by the least square method, and the system parameters of the plurality of first pressure sensors and the plurality of second pressure sensors are adjusted according to the plurality of pressure calibration coefficients; The displacement generator in the in-situ calibration module is used to apply a plurality of standard displacements to the plurality of vertical displacement sensors and the plurality of horizontal displacement sensors respectively, so as to obtain a plurality of displacement response data; The plurality of displacement response data are compared with the standard displacements respectively, a plurality of displacement calibration coefficients are calculated by the least square method, and the system parameters of the plurality of vertical displacement sensors and the plurality of horizontal displacement sensors are adjusted according to the plurality of displacement calibration coefficients.

8. The intelligent support system monitoring method of claim 6, wherein, The step of processing the initial sensor data set by the data processing module to obtain the support eccentricity, the actual rotation center and the performance degradation rate comprises: The data processing module is used for performing sliding average filtering and wavelet denoising on the initial sensor data set and removing abnormal values, so as to obtain a monitoring sensor data set, wherein the monitoring sensor data set comprises monitoring temperature data, monitoring pressure data, monitoring displacement data and monitoring rotation angle data; The monitoring temperature data is used for performing temperature compensation calculation on the monitoring pressure data, the monitoring displacement data and the monitoring rotation angle data, so as to obtain compensated pressure data, compensated displacement data and compensated rotation angle data; Based on the compensated pressure data, a support eccentricity is extracted, based on the compensated displacement data and the compensated rotation angle data, an actual rotation center is extracted, historical monitoring data is obtained, and based on the historical monitoring data, the compensated pressure data, the compensated displacement data and the compensated rotation angle data, a performance degradation rate is obtained.

9. The intelligent support system monitoring method of claim 6, wherein, The data processing module is used for constructing a membership cloud model, the support eccentricity, the actual rotation center and the performance degradation rate are input into the membership cloud model, and a monitoring evaluation result of the support is obtained. The data processing module is used for obtaining a training data set and a hyper-entropy coefficient, based on the training data set and the hyper-entropy coefficient, a mean method is used to determine a cloud model expectation and a cloud model hyper-entropy, so as to construct a cloud generator, the cloud generator comprises a cloud membership function, and the cloud membership function is used for calculating a cloud membership degree; A fuzzy neural network is obtained, the cloud membership is taken as an input of the fuzzy neural network, so as to compose the cloud generator and the fuzzy neural network into a membership cloud model; The support eccentricity, the actual rotation center and the performance degradation rate are input into the membership cloud model, and a monitoring evaluation result of the support is obtained.

10. The intelligent support system monitoring method of claim 9, wherein, The cloud membership function is: wherein, denotes cloud membership, denotes input value, denotes cloud drop number; wherein, represents a normal random membership, represents a cloud model expectation, represents a first normal random number calculated from the cloud model expectation and the cloud model hyper entropy as the expectation and the variance, respectively. represents a second normal random number calculated from the cloud model expectation and the cloud model hyper entropy as the expectation and the variance, respectively.

Images