Autonomous learning type rolling bearing fault identification and detection equipment
The self-learning rolling bearing fault identification and detection equipment utilizes a drive device and deep learning components for automated fault identification, solving the problems of long detection cycles and large errors in existing technologies, and achieving efficient and accurate rolling bearing detection.
Patent Information
- Application Number
- CN202423264579.0
- Authority / Receiving Office
- CN · China
- Patent Type
- Utility models(China)
- Current Assignee / Owner
- Filing Date
- 2024-12-30
- Publication Date
- 2025-11-18
- Estimated Expiration
- 2034-12-30
AI Technical Summary
Existing rolling bearing testing equipment has a long testing cycle, high cost, and relies on manual evaluation which is prone to errors, and it cannot learn abnormal conditions autonomously.
The self-learning rolling bearing fault identification and detection equipment includes a drive unit, a data acquisition unit, a simulated force application unit, and a detection and control unit. It uses deep learning components to identify faults through a two-dimensional convolutional neural network and integrates data preprocessing, deep learning, and fault identification components in the control box for automated detection.
It enables automated and rapid identification of rolling bearing faults, reduces human error, improves detection accuracy and range, and lowers detection costs.
Smart Images
Figure CN223565244U_ABST
Abstract
Description
TECHNICAL FIELD
[0001] The utility model relates to a rolling bearing detection equipment, especially to a self -learning type rolling bearing fault identification detection equipment. BACKGROUND
[0002] For the prior art, the rolling bearing is a commonly used transmission component, and its application field is wide. Therefore, the reliability of its operation will affect the stability of the equipment and instrument. Therefore, during the rolling bearing processing and maintenance, a plurality of detections need to be carried out to find possible abnormalities.
[0003] At present, the common abnormality is abnormal vibration during transmission. The conventional detection equipment is through the vibration collection end head to connect the transmission shaft of the rolling bearing to carry out data collection. Then, the data is judged in an artificial way, and when the data is in the abnormal range, it is determined that there is an abnormality. Then, it will be sent to other detection equipment for further detection. Therefore, the detection cycle is long, and different detection equipment needs to be selected according to different abnormalities, and the implementation cost is also high. And relying on artificial way, errors are easy to produce.
[0004] At the same time, there are also comparison equipment introduced to participate in the judgment of data. However, it needs to constantly update the database and passively input to improve the detection accuracy. It cannot independently collect and learn the possible abnormal state according to the frequently detected rolling bearing.
[0005] In view of the above defects, the designer actively researches and innovates to create a self -learning type rolling bearing fault identification detection equipment, so that it has more industrial utilization value. CONTENT OF THE UTILITY MODEL
[0006] To solve the above technical problems, the utility model aims at providing a self -learning type rolling bearing fault identification detection equipment.
[0007] The self -learning type rolling bearing fault identification detection equipment of the utility model, including equipment body, be provided with installation platform on equipment body, wherein: drive arrangement is installed on installation platform, drive arrangement is connected with test shaft through coupling assembly, one end of test shaft is installed with data acquisition device, test shaft is installed with positioning support, positioning support is installed with analog force applying device, detection control device is installed in equipment body, detection control device is connected with drive arrangement, data acquisition device, analog force applying device electricity, control device is installed on equipment body, control device is connected with detection control device.
[0008] Further, the self-learning type rolling bearing fault identification detection device, wherein the driving device is a motor, the driving end of the motor is connected with a shaft coupling assembly, the shaft coupling assembly is a shaft coupling, and the motor is connected with the mounting platform through a buffer support.
[0009] Further, the self-learning type rolling bearing fault identification detection device, wherein the test shaft comprises a shaft body, and a gasket is arranged outside the shaft body.
[0010] Further, the self-learning type rolling bearing fault identification detection device, wherein the data acquisition device comprises a mounting bracket connected with the mounting platform, a storage box is mounted outside the mounting bracket, a vibration sensor is mounted in the storage box, the vibration sensor is provided with a lead sleeve connected with the test shaft.
[0011] Further, the self-learning type rolling bearing fault identification detection device, wherein the positioning support is a gantry, mounting fins are extended from the lower end of the gantry, and the mounting fins are connected with the mounting platform through positioning screws.
[0012] Further, the self-learning type rolling bearing fault identification detection device, wherein the simulation force applying device comprises a servo motor, a driving shaft of the servo motor penetrates through the positioning support, and a pressure block is connected with the driving shaft.
[0013] Further, the self-learning type rolling bearing fault identification detection device, wherein the pressure block is a rectangular silica gel block, and circular arc contact surfaces are distributed at the lower end of the rectangular silica gel block.
[0014] Further, the self-learning type rolling bearing fault identification detection device, wherein the detection control device comprises an integrated control box, wiring terminals are arranged on the integrated control box, the wiring terminals are connected with the driving device, the data acquisition device, the simulation force applying device and the control device, a data bus is connected with the wiring terminals, a data preprocessing assembly connected with the data bus is mounted in the integrated control box, a deep learning assembly is connected with the output end of the data preprocessing assembly, a fault identification assembly is connected with the output end of the deep learning assembly, and a data storage assembly is also mounted in the integrated control box and connected with the data storage interfaces of the data preprocessing assembly, the deep learning assembly and the fault identification assembly.
[0015] Further, the self-learning type rolling bearing fault identification detection device, wherein the data preprocessing assembly is a vibration acceleration signal preprocessing assembly, the deep learning assembly is a processing assembly with a two-dimensional convolutional neural network model, the fault identification assembly is a fault signal reasoning assembly, and the data storage assembly is a solid state disk.
[0016] Further, the autonomous learning type rolling bearing fault identification detection device, wherein the operation device is a touch screen, and an indicator light is further installed on the device body and connected with the detection control device.
[0017] By the above scheme, the utility model has at least the following advantages:
[0018] 1. The movable overdrive device cooperates with the test shaft to provide rotating power for the rolling bearing for detection, without manual force applied to the rolling bearing, avoiding abnormal data caused by manual operation.
[0019] 2. The analog force applying device is provided to facilitate the rolling bearing to simulate common abnormal states, facilitate the detection control device to record and learn possible abnormalities, and realize autonomous fault reasoning and improve the detection range.
[0020] 3. The detection control device is provided to use a two-dimensional convolutional neural network to infer and judge possible fault signals through deep learning, improving the detection accuracy.
[0021] The above description is only a summary of the technical scheme of the utility model, in order to more clearly understand the technical means of the utility model, and the content of the specification can be implemented, the following is a preferred embodiment of the utility model and the detailed description of the drawings as follows. BRIEF DESCRIPTION OF DRAWINGS
[0022] Fig. 1 It is the overall structure schematic diagram of the autonomous learning type rolling bearing fault identification detection device.
[0023] Fig. 2 It is the cooperation schematic diagram of the analog force applying device and the rolling bearing.
[0024] Fig. 3 It is the configuration principle schematic diagram of the detection control device.
[0025] The meanings of the various reference signs in the drawings are as follows.
[0026] 1 device body 2 installation platform
[0027] 3 drive device 4 shaft coupling assembly
[0028] 5 test shaft 6 data acquisition device
[0029] 7 positioning support 8 analog force applying device
[0030] 9 detection control device 10 operation device
[0031] 11 shaft body 12 gasket
[0032] 13 mounting fin 14 pressing block
[0033] 15 arcuate contact surface 16 terminal
[0034] 17 data preprocessing component 18 deep learning component
[0035] 19 fault identification component 20 data storage component
[0036] 21 rolling bearing 22 indicator light DETAILED DESCRIPTION
[0037] The specific embodiments of the present application will be further described in detail below in conjunction with the accompanying drawings and examples. The following examples are used to illustrate the present application, but are not used to limit the scope of the present application.
[0038] As Figs. 1 to 3 The autonomous learning type rolling bearing fault identification detection equipment comprises an equipment body 1, and a mounting platform 2 is arranged on the equipment body 1. Different from the prior art, a driving device 3 is arranged on the mounting platform 2, and the driving device 3 is connected with a test shaft 5 through a shaft coupling assembly 4. During use, the rolling bearing 21 can be arranged on the test shaft 5 for steering driving, so as to simulate different working conditions. Meanwhile, a data acquisition device 6 is arranged at one end of the test shaft 5, so as to obtain the working condition of the rolling bearing 21 fed back to the test shaft 5. In addition, a positioning support 7 is arranged outside the test shaft 5, and a simulation force applying device 8 is arranged on the positioning support 7. In this way, the simulation force applying device 8 can provide an external force to the rolling bearing 21 to simulate an abnormality and meet the expansion of the detection project. Furthermore, in consideration of the convenience of data processing, a detection control device 9 is arranged in the equipment body 1, and the detection control device 9 is electrically connected with the driving device 3, the data acquisition device 6 and the simulation force applying device 8. A control device 10 is arranged on the equipment body 1 and connected with the detection control device 9, so as to realize integrated control. A power supply interface and a power supply wire are prearranged in the equipment body 1, and the power supply wire is connected with each device.
[0039] In a preferred embodiment of the present application, the driving device 3 is a motor, the driving end of the motor is connected with the shaft coupling assembly 4, the shaft coupling assembly 4 is a shaft coupling, and the motor is connected with the mounting platform 2 through a buffer support. Meanwhile, the test shaft 5 comprises a shaft body 11, and a gasket 12 is arranged outside the shaft body 11. In this way, the gasket 12 suitable for the inner diameter model of the rolling bearing 21 can be selected to ensure that the gasket 12 is stably connected with the test shaft 5. In this way, data anomalies caused by loosening of the joint end surface can be avoided.
[0040] Further, the data acquisition device 6 comprises a mounting bracket connected with the mounting platform 2, a receiving box is mounted outside the mounting bracket, a vibration sensor is mounted in the receiving box, the vibration sensor is provided with a lead-through sleeve connected with the test shaft 5. In this way, the abnormal vibration caused by the transmission shaft can be effectively collected to perceive the fault state of the rolling bearing 21.
[0041] In combination with the actual implementation, the positioning support 7 is a gantry, the lower end of the gantry extends a mounting fin 13 connected with the mounting platform 2 through a positioning screw. At the same time, the analog force applying device 8 comprises a servo motor, the driving shaft of the servo motor penetrates through the positioning support 7, and a pressure block 14 is connected on the driving shaft. Moreover, the pressure block 14 is a rectangular silica gel block, and the lower end of the rectangular silica gel block is distributed with a circular arc contact surface 15. In this way, the external stress transmitted to the rolling bearing 21 through the pressure block 14 can simulate various fault abnormalities of the rolling bearing 21, so that various abnormal vibrations can be learned and identified through the detection control device 9 in the early detection stage, and more accurate comparison can be realized in the subsequent detection period.
[0042] Further, the detection control device 9 comprises an integrated control box, a wiring terminal 16 is arranged on the integrated control box, and the wiring terminal 16 is connected with the driving device 3, the data acquisition device 6, the analog force applying device 8 and the control device 10. Specifically, the wiring terminal 16 is connected with a data bus, and a data preprocessing component 17 connected with the data bus is mounted in the integrated control box. At the same time, the output end of the data preprocessing component 17 is connected with a deep learning component 18, and the output end of the deep learning component 18 is connected with a fault identification component 19. In this way, through the training and learning in the early stage, the detection control device 9 can effectively identify various possible abnormalities, and can continuously accumulate and compare abnormal values to improve the angle of subsequent detection and identification. In addition, considering the optimization of data storage and retrieval, a data storage component 20 is also mounted in the integrated control box, and the data storage component 20 is respectively connected with the data storage interfaces of the data preprocessing component 17, the deep learning component 18 and the fault identification component 19.
[0043] During implementation, in order to effectively realize the self-learning of the detection data, the data preprocessing component 17 is a vibration acceleration signal preprocessing component, and the deep learning component 18 is a processing component with a two-dimensional convolutional neural network model. Meanwhile, the fault identification component 19 is a fault signal reasoning component, and the data storage component 20 is a solid state disk. In this way, through the preset abnormality discrimination program, the acquired vibration data are used to perform reasoning discrimination on the corresponding abnormal state. Furthermore, according to the use requirement, the corresponding program can be pre-recorded in the deep learning component 18 and the fault identification component 19. The data preprocessing component 17, the deep learning component 18 and the fault identification component 19 can also be directly integrated according to the MCU mode. The program and the integrated configuration mode are not the object of protection of the utility model, and will not be described here.
[0044] Considering the convenience of user operation, the pre-stored program and other functions can be conveniently switched by point selection. The operation device 10 is a touch screen. Meanwhile, the device body 1 is also provided with an indicator lamp 22 connected with the detection control device 9. In this way, the current detection state can be more intuitively obtained, for example, green for normal and red for abnormal, so as to facilitate the user to quickly intervene in processing.
[0045] The working principle of the utility model is as follows:
[0046] Self-learning stage
[0047] The normal or abnormal rolling bearing 21 is installed on the test shaft 5 to complete the docking of the data acquisition device 6.
[0048] Then, the rolling bearing 21 is driven to rotate along with the operation of the driving device 3, and the data acquisition device 6 collects data. The detection control device 9 learns and discriminates the abnormality according to the preset program.
[0049] During this period, the simulated force applying device 8 can be controlled to apply different stresses to the rolling bearing 21 to simulate other abnormalities. The detection control device 9 can still learn the various abnormalities that the rolling bearing 21 can have.
[0050] Actual detection stage
[0051] The rolling bearing 21 to be detected is installed on the test shaft 5 to complete the docking of the data acquisition device 6. Then, the rolling bearing 21 is driven to rotate along with the operation of the driving device 3, and the data acquisition device 6 collects data.
[0052] Subsequently, the detection control device 9 compares and calculates the current data, and outputs the collected data and the comparison result through the touch screen. If it is directly determined that there is an abnormality, the indicator lamp 22 can directly alarm.
[0053] Thus, it is convenient for the user to participate in the disposal.
[0054] As can be seen from the above description and in combination with the drawings, the present application has the following advantages:
[0055] 1. The movable overdrive device cooperates with the test shaft to provide rotating power for the rolling bearing, without manual force on the rolling bearing, avoiding abnormal data caused by manual operation.
[0056] 2. The simulation force device is provided, which can simulate common abnormal states of the rolling bearing, facilitate the detection control device to record and learn possible abnormalities, and realize autonomous fault reasoning and improve the detection range.
[0057] 3. The detection control device is provided, which can use a two-dimensional convolutional neural network to infer and judge possible fault signals through deep learning, thereby improving the detection accuracy.
[0058] In addition, the indication direction or position relationship described in the present application is based on the direction or position relationship shown in the drawings, which is only for the convenience of describing the present application and simplifying the description, and does not indicate or imply that the device or structure must have a specific direction or be operated in a specific direction, so it cannot be understood as a limitation of the present application.
[0059] The above description is only the preferred embodiment of the present application, and is not used to limit the present application. It should be pointed out that for ordinary skilled persons in the art, without departing from the technical principles of the present application, a number of improvements and modifications can be made, and these improvements and modifications should be regarded as the protection scope of the present application.
Claims
1. A self-learning rolling bearing fault identification and detection device, comprising a device body, wherein an installation platform is provided on the device body, characterized in that: A drive unit is installed on the mounting platform. The drive unit is connected to a test shaft via a coupling assembly. A data acquisition device is installed at one end of the test shaft. A positioning bracket is installed outside the test shaft. A simulated force application device is installed on the positioning bracket. A detection and control device is installed inside the equipment body. The detection and control device is electrically connected to the drive unit, the data acquisition device, and the simulated force application device. A control device is installed on the equipment body and is connected to the detection and control device.
2. The self-learning rolling bearing fault identification and detection device according to claim 1, characterized in that: The driving device is a motor, the driving end of the motor is connected to a coupling assembly, the coupling assembly is a coupling, and the motor is connected to the mounting platform through a buffer support.
3. The self-learning rolling bearing fault identification and detection device according to claim 1, characterized in that: The test shaft includes a shaft body, and a washer is fitted on the outer side of the shaft body.
4. The self-learning rolling bearing fault identification and detection device according to claim 1, characterized in that: The data acquisition device includes a mounting frame connected to the mounting platform, a storage box mounted outside the mounting frame, a vibration sensor installed inside the storage box, and a guide sleeve configured for the vibration sensor, which is connected to the test shaft.
5. The self-learning rolling bearing fault identification and detection device according to claim 1, characterized in that: The positioning bracket is a gantry frame, and the lower end of the gantry frame extends with mounting fins. The mounting fins are connected to the mounting platform by isomorphic positioning screws.
6. The self-learning rolling bearing fault identification and detection device according to claim 1, characterized in that: The simulated force application device includes a servo motor, the drive shaft of which passes through a positioning bracket, and a pressure block is connected to the drive shaft.
7. The self-learning rolling bearing fault identification and detection device according to claim 6, characterized in that: The pressure block is a rectangular silicone block, and the lower end of the rectangular silicone block has an arc-shaped contact surface.
8. The self-learning rolling bearing fault identification and detection device according to claim 1, characterized in that: The detection and control device includes an integrated control box with terminal blocks connected to a drive device, a data acquisition device, a simulated force application device, and a control device. A data bus is also connected to the terminal blocks. A data preprocessing component connected to the data bus is installed inside the integrated control box. A deep learning component is connected to the output of the data preprocessing component, and a fault identification component is connected to the output of the deep learning component. A data storage component is also installed inside the integrated control box, and it interfaces with the data storage interfaces of the data preprocessing component, the deep learning component, and the fault identification component.
9. The self-learning rolling bearing fault identification and detection device according to claim 8, characterized in that: The data preprocessing component is a vibration acceleration signal preprocessing component, the deep learning component is a processing component with a two-dimensional convolutional neural network model, the fault identification component is a fault signal inference component, and the data storage component is a solid-state drive.
10. The self-learning rolling bearing fault identification and detection device according to claim 1, characterized in that: The control device is a touch screen, and the device body is also equipped with indicator lights, which are connected to the detection and control device.