Joint bearing torque monitoring device and method based on optical fiber sensing

Abstract

The present application relates to the technical field of structural health monitoring, and particularly relates to a joint bearing torque monitoring device and a monitoring method based on optical fiber sensing. The joint bearing torque monitoring device based on optical fiber sensing provided by the present application can reverse the torque when the joint bearing rotates, the swing direction and the bending moment when the joint bearing swings based on the optical fiber grating sensor layout scheme, and can realize real-time monitoring of the torque of the joint bearing under the service load condition of the joint bearing by cooperating with the optical fiber sensing acquisition equipment and the cloud platform. The joint bearing torque monitoring method based on optical fiber sensing provided by the present application overcomes the characteristics that the strain gauge sensor is easy to be disturbed by electromagnetic interference, and different analysis methods can be selected based on the same set of devices for different motion forms, so as to obtain accurate and timely abnormal early warning information, so as to achieve the purpose of condition-based maintenance. In addition, the real-time output and storage of the abnormal early warning information can also be used for long-term strain trend analysis, and the maintenance time can be predicted in advance, so as to reduce the potential risk.

Description

Technical Field

[0001] This invention relates to the field of structural health monitoring technology, and in particular to a joint bearing torque monitoring device and method based on fiber optic sensing. Background Technology

[0002] Spherical plain bearings are widely used in the service environments of major equipment in aerospace, machinery manufacturing and other fields, where they play a role in motion and support. Their health status during service is of great importance to the safe and stable operation of major equipment.

[0003] A spherical plain bearing is a special type of bearing consisting of an outer ring with an inner spherical surface and an inner ring with an outer spherical surface, capable of rotation, oscillation, and combinations thereof. During the relative movement of the inner and outer spherical surfaces, corresponding frictional resistance must be overcome. If this frictional resistance increases abnormally, it may lead to impeded movement or even jamming. Since changes in frictional resistance directly affect the operating condition of the spherical plain bearing, monitoring it is crucial. Typically, the magnitude of frictional resistance can be characterized by frictional torque, thereby quantifying the health status of the spherical plain bearing and providing a basis for early fault prediction.

[0004] Currently, the measurement of friction torque in spherical plain bearings mainly relies on indirect detection using external measuring devices. For example, the self-lubricating spherical plain bearing friction torque detection device disclosed in Chinese Patent CN209296314U on August 23, 2019, uses a torque box connected to a fixed support shaft with an axially connected angular displacement sensor. A loading module is connected to the torque box containing the test specimen. The bearing swing module transmits rotational torque to the torque box. The loading module simulates the working conditions of the central shaft of the test specimen according to the instructions of the industrial control system, measures the torque value during the swing process of the torque box, and feeds the torque value back to the industrial control system. The industrial control system calculates the friction torque of the detection device itself and the friction coefficient of the radial ball bearing in the test specimen. Then, the test specimen containing the radial ball bearing in the torque box is replaced with a self-lubricating spherical plain bearing. The loading module loads according to the instructions of the industrial control system, measures the torque during the swing process, and calculates the friction torque of the self-lubricating spherical plain bearing in the test specimen through the industrial control system.

[0005] The method described above still derives the magnitude of the frictional torque by measuring the response of the external load and the bearing structure. However, its measurement path is complex and still insufficient to meet the needs of online monitoring. Therefore, how to achieve real-time monitoring of torque changes in spherical plain bearings during service, overcoming the limitations of traditional methods, remains a problem that those skilled in the art need to solve. Summary of the Invention

[0006] To address the technical problem of real-time monitoring of torque changes in spherical plain bearings during service, this invention provides a spherical plain bearing torque monitoring device based on fiber optic sensing, comprising: A fiber optic strain sensor is mounted on a drive shaft connected to the inner ring of a spherical bearing for torque monitoring. A fiber optic temperature sensor is positioned near the fiber optic strain sensor to provide temperature compensation during the torque monitoring process of the fiber optic strain sensor. Fiber optic sensing and acquisition equipment is used to acquire, store, and transmit the changes in the center wavelength of the reflection spectrum of fiber optic strain sensors and fiber optic temperature sensors in real time. The cloud platform is used to process and analyze the collected sensor data.

[0007] In one embodiment, the fiber Bragg grating temperature sensor is obtained by encapsulating the fiber Bragg grating strain sensor in a copper tube.

[0008] In one embodiment, four fiber optic strain sensors are arranged at 0°, 90°, 180° and 270° along the circumference of the drive shaft connected to the inner ring of the spherical bearing.

[0009] Furthermore, the angle α between the axial direction of the fiber optic strain sensor and the axial direction of the drive shaft is 30° to 45°.

[0010] The present invention also provides a method for monitoring the torque of a joint bearing based on fiber optic sensing, which, based on the fiber optic sensing-based joint bearing torque monitoring device described above, includes the following steps: S100, deploy sensors, debug fiber optic sensing and acquisition equipment, and deploy cloud platform; S200: Acquires center wavelength data of the reflection spectrum of fiber optic strain sensor and fiber optic temperature sensor, and performs temperature compensation analysis and filtering noise reduction. S300: Monitor and analyze the motion of the spherical bearing; if the strain value is abnormal, output an early warning message.

[0011] In one embodiment, step S100 includes: S110. Sequentially install fiber optic strain sensors and fiber optic temperature sensors; S120, preheating fiber optic sensing and acquisition equipment, stabilizing acquisition level; S130. Build an artificial neural network model and complete model training and evaluation verification; S140. Deploy the artificial neural network model to the cloud platform and set early warning values.

[0012] Furthermore, step S130 includes: S131. The fiber optic sensing and acquisition equipment collects the change in the center wavelength of the reflection spectrum of the fiber optic strain sensor and the fiber optic temperature sensor during the rotation of the joint bearing and transmits it to the cloud platform. S132. After temperature compensation using the dual grating method, the cloud platform eliminates the influence of wavelength drift caused by temperature, then uses Kalman filtering for noise reduction, and finally calculates the conversion coefficient to obtain the strain data at the measurement point of the fiber optic strain sensor. S133. Normalize the strain data and use it as input to the artificial neural network model; S134. Divide the collected sensor data into a training sample set and a test sample set according to the proportion; S135. Train the artificial neural network model based on the training sample set; S136. The artificial neural network model is verified and evaluated using the test sample set.

[0013] Furthermore, the training sample set and the test sample set are sample sets containing several sets of normalized fiber grating strain values ​​and their corresponding stress moment labels. The torque label is derived from the strain data obtained in step S132 using the following simplified formula.

[0014]

[0015] In the formula, ν α is Poisson's ratio, and α is the fiber Bragg grating sensor placement angle. This represents the strain value of the fiber optic grating. E The elastic modulus of the material. τ For shear stress, J It is the polar moment of inertia. R The outer radius of the drive shaft. T For torque.

[0016] In one embodiment, when the joint bearing's motion is rotational, step S300 specifically involves: S311. Input the strain data obtained after temperature compensation and filtering noise reduction into the artificial neural network model; S312, The artificial neural network model outputs the predicted friction torque value; S313. If the cloud platform determines that the friction torque value is higher than the set warning value for rotational motion, it will output a warning message.

[0017] In one embodiment, when the joint bearing's motion is a swinging motion, step S300 specifically involves: S311. Based on the strain data obtained after temperature compensation and filtering noise reduction, construct symmetrical differential strain in the x and y directions, and determine the main direction of oscillation; S312. Perform moment synthesis calculations in the x and y directions to obtain bending moment values, and calculate the swing direction angle to obtain swing information; S313. If the cloud platform determines that the swing information does not match the set command, it will output a warning message.

[0018] In summary, compared with the prior art, the invention has the following beneficial effects: The spherical bearing torque monitoring device based on fiber optic sensing provided by this invention deploys a suitable number of fiber optic grating sensors at a specific angle on the drive shaft arm connected to the inner ring of the spherical bearing. By using the fiber optic grating sensors in conjunction with fiber optic sensing acquisition equipment and a cloud platform, stress and temperature data during service can be obtained and analyzed, thereby realizing real-time monitoring of the torque of the spherical bearing under service load.

[0019] Compared to strain gauge sensors, the fiber optic grating sensor used in this invention has the advantage of being resistant to electromagnetic interference; compared to solutions using external torque detection devices, this invention can realize online monitoring of friction torque of spherical bearings during service, providing a reliable basis for health assessment and timely maintenance of spherical bearings, and reducing potential risks.

[0020] The method for monitoring the torque of a joint bearing based on fiber optic sensing provided by this invention realizes the inversion of the magnitude of the torque when the joint bearing rotates and the inversion of the swing direction and bending moment when it swings through the deployment scheme; and overcomes the cross-sensitivity of temperature strain of fiber optic grating sensors by using the dual-grating temperature compensation method.

[0021] Specifically, the temperature-compensated stress data is filtered using a Kalman filter to reduce the impact of data noise, and then differentiated and customized analyses are performed based on the operating conditions. When the spherical bearing rotates, the compensated and noise-reduced stress data is input into an artificial neural network model to predict the torque information during rotation; when the spherical bearing oscillates, the obtained oscillation information is analyzed to determine whether the oscillation direction and amplitude are abnormal. For different motion forms, based on the same device, different analysis methods are selected to obtain accurate and timely anomaly warning information, thereby achieving the purpose of condition-based maintenance. Among them, the dual-grating temperature compensation and Kalman filtering work together to eliminate temperature environment interference and further reduce random noise. At the same time, the symmetrical layout of the four sensors and differential analysis can simultaneously collect strain in the x / y direction, providing a basis for torque inversion of rotational motion and direction determination of oscillating motion. This allows the same set of hardware to adapt to two motion forms, avoiding the cumbersome operation of replacing sensors or adjusting the layout required by existing technologies.

[0022] In addition, the real-time output and storage of abnormal early warning information can also be used for long-term strain trend analysis to determine whether there is fatigue loss in the joint bearing, thereby enabling early prediction of maintenance time.

[0023] Other features and beneficial effects of the invention will be set forth in the following description, and will be apparent in part from the description, or may be learned by practicing the invention. The objects and other beneficial effects of the invention can be realized and obtained by means of the structures particularly pointed out in the description, claims and drawings. Attached Figure Description

[0024] To more clearly illustrate the technical solutions in the embodiments of the present invention or the prior art, the drawings used in the description of the embodiments or the prior art will be briefly introduced below. Obviously, the drawings described below are some embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on these drawings without creative effort.

[0025] Figure 1 This is a schematic diagram of the strain FBG sensor and temperature FBG sensor provided in Embodiment 1 of the present invention; Figure 2 This is a schematic diagram of the strain FBG sensor layout provided in Embodiment 1 of the present invention; Figure 3 This is a schematic diagram of the workflow of the joint bearing torque monitoring method based on fiber optic sensing provided in Embodiment 2 of the present invention. Figure 4 This is a diagram illustrating the rotational motion analysis of a joint bearing provided in Embodiment 2 of the present invention. Figure 5 This is a diagram illustrating the oscillating motion analysis of a joint bearing provided in Embodiment 3 of the present invention.

[0026] Figure label: 100 - Fiber Bragg grating sensor; 110 - Strain FBG sensor; 111 - First optical cable; 120 - Temperature FBG sensor; 121 - Second optical cable; 122 - Copper tube; 210 - Outer ring; 220 - Inner ring; 230 - Drive shaft. Detailed Implementation

[0027] To make the objectives, technical solutions, and advantages of the embodiments of the present invention clearer, the technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0028] In the description of this invention, it should be noted that the terms "upper," "lower," "inner," and "outer," etc., indicate the orientation or positional relationship based on the orientation or positional relationship shown in the accompanying drawings. They are used only for the convenience of describing the invention and for simplifying the description, and do not indicate or imply that the device or element referred to must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, they should not be construed as limitations on the invention. Furthermore, the terms "first" and "second" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0029] Example 1 This embodiment provides a joint bearing torque monitoring device based on fiber optic sensing, including a fiber optic grating sensor 100, a fiber optic sensing acquisition device, and a cloud platform. The fiber optic grating sensor 100 includes a fiber optic grating strain sensor and a fiber optic grating temperature sensor, and the fiber optic grating sensor 100 is connected to the fiber optic sensing acquisition device, which is connected to the cloud platform. A fiber optic strain sensor is mounted on a drive shaft connected to the inner ring of a spherical bearing for torque monitoring. A fiber optic temperature sensor is positioned near the fiber optic strain sensor to provide temperature compensation during the torque monitoring process of the fiber optic strain sensor. Fiber optic sensing and acquisition equipment is used to acquire, store, and transmit the changes in the center wavelength of the reflection spectrum of fiber optic strain sensors and fiber optic temperature sensors in real time. The cloud platform is used to process and analyze the collected sensor data.

[0030] See Figure 1 As shown, the fiber optic strain sensor is a strain FBG sensor 110, and the fiber optic temperature sensor is a temperature FBG sensor 120, which is obtained by encapsulating the strain FBG sensor 110 in a copper tube 122.

[0031] See Figure 2 As shown, there are four strain FBG sensors 110, which are arranged at 0°, 90°, 180° and 270° along the circumference of the transmission shaft 230 connected to the inner ring 220 of the spherical bearing. In this embodiment, the angle α between the axial direction of the sensor and the axial direction of the transmission shaft is 30°, that is, the arrangement angle α is 30°. In specific implementation, several strain FBG sensors 110 are connected in series through a first optical cable 111 and finally connected to an optical fiber sensing and acquisition device; several temperature FBG sensors 120 are connected in series through a second optical cable 121 and finally connected to an optical fiber sensing and acquisition device; the strain FBG sensors 110 and temperature FBG sensors 120 are both adhered and deployed using epoxy resin or other fixing materials.

[0032] Example 2 Based on the fiber optic sensing-based joint bearing torque monitoring device described in Embodiment 1 of the present invention, see [reference]. Figure 3 As shown, this embodiment provides a method for monitoring the torque of a joint bearing based on fiber optic sensing, which monitors the torque changes of the joint bearing in real time during rotational motion. The method specifically includes the following steps: S100, deploy sensors, debug fiber optic sensing and acquisition equipment, and deploy cloud platform; S200: Acquires center wavelength data of the reflection spectrum of fiber optic strain sensor and fiber optic temperature sensor, and performs temperature compensation analysis and filtering noise reduction. S300: Monitor and analyze the motion of the spherical bearing; if the strain value is abnormal, output an early warning message.

[0033] Furthermore, step S100 includes: S110, sequentially deploy fiber optic strain sensors and fiber optic temperature sensors. S120, preheating fiber optic sensing and acquisition equipment, stabilizing acquisition level; S130. Build an artificial neural network model and complete model training and evaluation verification; S140. Deploy the artificial neural network model to the cloud platform and set early warning values.

[0034] For specific implementation steps S110, please refer to... Figure 4 As shown, fiber optic grating sensors 100 are arranged in 0°, 90°, 180° and 270° orientations along the circumference of the drive shaft 230 connected to the inner ring 220 of the spherical bearing using epoxy resin, with an arrangement angle α of 30°.

[0035] Preferably, the fiber optic grating sensor 100 should be selected from the same batch to maintain the consistency of the sensor's strain and temperature sensitivity coefficients; more preferably, the epoxy resin is a low-modulus epoxy resin, and the curing condition is to cure at room temperature for 24 hours to ensure that the sensor and the drive shaft are rigidly connected and to avoid measurement errors caused by relative deformation.

[0036] In specific implementation step S120, after the level stabilizes, the initial Bragg wavelength of the fiber optic grating sensor 100 is acquired.

[0037] Furthermore, step S130 includes: S131. The fiber optic sensing and acquisition equipment collects the change in the center wavelength of the reflection spectrum of the fiber optic strain sensor and the fiber optic temperature sensor during the rotation of the joint bearing and transmits it to the cloud platform. S132. After temperature compensation using the dual grating method, the cloud platform eliminates the influence of wavelength drift caused by temperature, then uses Kalman filtering for noise reduction, and finally calculates the conversion coefficient to obtain the strain data at the measurement point of the fiber optic strain sensor. S133. Normalize the strain data and use it as input to the artificial neural network model; S134. The collected sensor data is divided into a training sample set and a test sample set according to the proportion. S135. Train the artificial neural network model based on the training sample set; S136. The artificial neural network model is verified and evaluated using the test sample set.

[0038] In specific implementation step S132, the fiber grating is sensitive to both strain and temperature, and the relationship between them can be expressed as shown in equation [1], where, For total wavelength drift, Wavelength shift caused by mechanical strain Wavelength shift caused by temperature;

[0039] In this embodiment of the invention, the temperature FBG sensor 120 is obtained by encapsulating the strain FBG sensor 110 in a copper tube 112, thus isolating it from the influence of deformation. If we consider it as 0, then we can obtain the ambient temperature at that moment based on the wavelength change of the temperature FBG sensor 120 and perform temperature compensation on the strain FBG sensor 110. Then, we use Kalman filtering to filter the temperature-compensated data to reduce noise. The wavelength change of the strain FBG sensor 110 after temperature compensation and noise reduction is denoted as... ; Furthermore, the strain data caused by mechanical deformation at the measurement point of the strain FBG sensor 110 under the current state is obtained by the absolute wavelength algorithm [2]. Wherein, K is the conversion coefficient, which is used to characterize the drift length of the Bragg wavelength when 1 microstrain is generated; it should be noted that in the embodiments of the present invention, a conversion coefficient of 1.2 pm / με is used.

[0040]

[0041] In specific implementation step S133, the strain data after temperature compensation is normalized according to formula [3], wherein, , The collected strain data are respectively The minimum and maximum values ​​in.

[0042]

[0043] In specific implementation step S134, multiple sets of data are collected under the same load conditions and in the state of rotational motion. The first 70% to 80% are divided into training sample set, and the last 20% to 30% are divided into test sample set. Furthermore, the training sample set and the test sample set are sample sets containing several sets of fiber grating strain values ​​and their corresponding stress moment labels; Specifically, based on strain data and according to equation [4] and the placement angle α, the shear stress at the measurement point of strain FBG sensor 110 can be obtained. τ The size, in formula [4], ν Poisson's ratio, E The elastic modulus of the material; based on the result of equation [4], the torque at the measurement point of the strain FBG sensor 110 can be obtained according to equation [5]. T The size, in formula [5], J It is the polar moment of inertia. R The outer radius of the drive shaft 230.

[0044]

[0045]

[0046] Preferably, multiple sets of torque values ​​under the same load are obtained, and the average value is taken as the torque label in the training sample set; then the load size is changed, and multiple sets of fiber optic strain values ​​and their corresponding stress torque labels are obtained after the load size is changed.

[0047] The device performs N rounds of data acquisition, with each round including four feature dimensions. , , , Each of the four feature dimensions corresponds to a torque information label, resulting in 4N samples, which together form a training sample set and a test sample set. In this embodiment, in the N rounds of data collection, the first 80% is used as the training sample set, and the last 20% is used as the test sample set.

[0048] In specific implementation step S135, it is preferable to simultaneously use the Adam algorithm to optimize the loss function to minimize the difference between the predicted value and the observed torque.

[0049] Furthermore, in this embodiment, step S300 is as follows: S311. Input the strain data obtained after temperature compensation and filtering noise reduction into the artificial neural network model; S312, The artificial neural network model outputs the predicted friction torque value; S313. If the cloud platform determines that the friction torque value is higher than the set warning value for rotational motion, it will output a warning message.

[0050] In practice, during the service of the spherical bearing, the fiber optic sensing acquisition device collects the wavelength information of the strain FBG sensor 110 and the temperature FBG sensor 120 in real time, transmits it to the cloud platform for analysis, calculates the strain data after temperature compensation and filtering noise reduction, inputs it into the artificial neural network model to obtain the predicted friction torque value, and if the cloud platform determines that it is higher than the set warning value for rotational motion, it outputs a warning message; the warning message includes an alarm prompt and the predicted friction torque value.

[0051] Example 3 Based on the fiber optic sensing-based joint bearing torque monitoring device described in Embodiment 1 of the present invention, this embodiment provides a fiber optic sensing-based joint bearing torque monitoring method to monitor the torque change of the joint bearing during swinging motion in real time, specifically including the following steps: S100, deploy sensors, debug fiber optic sensing and acquisition equipment, and deploy cloud platform; S200: Acquire center wavelength data of the reflection spectrum of fiber optic strain sensor and fiber optic temperature sensor, and perform temperature compensation analysis. S300: Monitor and analyze the motion of the spherical bearing; if the strain value is abnormal, output an early warning message.

[0052] Steps S100 and S200 are the same as in Embodiment 2 of the present invention, except that step S300 is specifically as follows: S311. Based on the strain data obtained after temperature compensation and filtering noise reduction, construct symmetrical differential strain in the x and y directions, and determine the main direction of oscillation; S312. Perform moment synthesis calculations in the x and y directions to obtain bending moment values, and calculate the swing direction angle to obtain swing information; S313. If the cloud platform determines that the swing information does not meet the set conditions, it will output a warning message.

[0053] See Figure 5 In this embodiment, when implementing step S311, symmetrical differential strains in the x and y directions are constructed according to equations [6] and [7], and then compared. , The size, if Then the main direction of the swing is the x-direction, that is, the bending is along the x-direction. If the main direction of the swing is the y-direction, then the bending is along the y-direction.

[0054]

[0055]

[0056] In this embodiment, when implementing step S312, the bending moment is calculated using equation [8]. M The size, in the formula, E The elastic modulus of the material. I Let the moment of inertia of the cross section be... r The distance from the neutral layer is used to calculate the swing direction angle using formula [9], thereby obtaining swing information containing bending moment and swing direction angle.

[0057]

[0058]

[0059] In this embodiment, when implementing step S313, the obtained swing information is compared with the set specification. If at least one item is inconsistent, a warning message is output. The warning message includes an alarm prompt and the bending moment and the swing direction angle.

[0060] Although this paper frequently uses terms such as fiber optic sensing, fiber optic strain sensor, fiber optic temperature sensor, fiber optic sensing acquisition device, cloud platform, fiber optic grating sensor, strain FBG sensor, temperature FBG sensor, spherical bearing, inner ring, outer ring, drive shaft, connecting shaft, temperature compensation, artificial neural network model, wavelength change, torque information, strain data, oscillation information, and early warning information, the possibility of using other terms is not excluded. These terms are used merely for the convenience of describing and explaining the essence of this invention; interpreting them as any additional limitation would contradict the spirit of this invention.

[0061] Finally, it should be noted that the above embodiments are only used to illustrate the technical solutions of the present invention, and not to limit them; although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some or all of the technical features; and these modifications or substitutions do not cause the essence of the corresponding technical solutions to deviate from the scope of the technical solutions of the embodiments of the present invention.

Claims

1. A joint bearing torque monitoring device based on fiber optic sensing, characterized in that, include: A fiber optic strain sensor is mounted on a drive shaft connected to the inner ring of a spherical bearing for torque monitoring. A fiber optic temperature sensor is positioned near the fiber optic strain sensor to provide temperature compensation during the torque monitoring process of the fiber optic strain sensor. Fiber optic sensing and acquisition equipment is used to acquire, store, and transmit the changes in the center wavelength of the reflection spectrum of fiber optic strain sensors and fiber optic temperature sensors in real time. The cloud platform is used to process and analyze the collected sensor data.

2. The joint bearing torque monitoring device based on fiber optic sensing according to claim 1, characterized in that: The fiber optic temperature sensor is obtained by encapsulating the fiber optic strain sensor in a copper tube.

3. The joint bearing torque monitoring device based on fiber optic sensing according to claim 2, characterized in that: The fiber optic strain sensor consists of four units, which are arranged in 0°, 90°, 180° and 270° orientations along the circumference of the drive shaft connected to the inner ring of the spherical bearing.

4. The joint bearing torque monitoring device based on fiber optic sensing according to claim 3, characterized in that: The angle α between the axis of the fiber optic strain sensor and the axis of the drive shaft is 30° to 45°.

5. A method for monitoring the torque of a joint bearing based on fiber optic sensing, wherein the joint bearing torque monitoring device based on fiber optic sensing as described in any one of claims 1 to 4 is characterized in that, Includes the following steps: S100, deploy sensors, debug fiber optic sensing and acquisition equipment, and deploy cloud platform; S200: Acquires center wavelength data of the reflection spectrum of fiber optic strain sensor and fiber optic temperature sensor, and performs temperature compensation analysis and filtering noise reduction. S300: Monitor and analyze the motion of the spherical bearing; if the strain value is abnormal, output an early warning message.

6. The method for monitoring the torque of a joint bearing based on fiber optic sensing according to claim 5, characterized in that, Step S100 includes: S110. Sequentially install fiber optic strain sensors and fiber optic temperature sensors; S120, preheated fiber optic sensing and acquisition equipment, stable acquisition level; S130. Build an artificial neural network model and complete model training and evaluation verification; S140. Deploy the artificial neural network model to the cloud platform and set early warning values.

7. The method for monitoring the torque of a joint bearing based on fiber optic sensing according to claim 6, characterized in that, Step S130 includes: S131. The fiber optic sensing and acquisition equipment collects the change in the center wavelength of the reflection spectrum of the fiber optic strain sensor and the fiber optic temperature sensor during the rotation of the joint bearing and transmits it to the cloud platform. S132. After temperature compensation using the dual grating method, the cloud platform eliminates the influence of wavelength drift caused by temperature, then uses Kalman filtering for noise reduction, and finally calculates the conversion coefficient to obtain the strain data at the measurement point of the fiber optic strain sensor. S133. Normalize the strain data and use it as input to the artificial neural network model; S134. Divide the collected sensor data into a training sample set and a test sample set according to the proportion; S135. Train the artificial neural network model based on the training sample set; S136. The artificial neural network model is verified and evaluated using the test sample set.

8. The method for monitoring the torque of a joint bearing based on fiber optic sensing according to claim 7, characterized in that: The training sample set and the test sample set are sample sets containing several sets of normalized fiber grating strain values ​​and their corresponding stress moment labels. The torque label is derived from the strain data obtained in step S132 using the following simplified formula. In the formula, ν α is Poisson's ratio, and α is the fiber Bragg grating sensor placement angle. This represents the strain value of the fiber optic grating. E The elastic modulus of the material. τ For shear stress, J It is the polar moment of inertia. R The outer radius of the drive shaft. T For torque.

9. The method for monitoring the torque of a joint bearing based on fiber optic sensing according to claim 8, characterized in that, When the joint bearing's motion is rotational, step S300 specifically involves: S311. Input the strain data obtained after temperature compensation and filtering noise reduction into the artificial neural network model; S312, The artificial neural network model outputs the predicted friction torque value; S313. If the cloud platform determines that the friction torque value is higher than the set warning value for rotational motion, it will output a warning message.

10. The method for monitoring the torque of a joint bearing based on fiber optic sensing according to claim 8, characterized in that, When the joint bearing's motion is a swinging motion, step S300 specifically involves: S311. Based on the strain data obtained after temperature compensation and filtering noise reduction, construct symmetrical differential strain in the x and y directions, and determine the main direction of oscillation; S312. Perform moment synthesis calculations in the x and y directions to obtain bending moment values, and calculate the swing direction angle to obtain swing information; S313. If the cloud platform determines that the swing information does not match the set command, it will output a warning message.

Images