A detection device and method for rolling element defects under operating conditions.

Abstract

This invention discloses a detection device and method for rolling element defects under operating conditions, comprising a track, rolling elements, a high-speed moving camera, a linear motion body, a rolling element tread cleaning mechanism, a rolling element image detection controller, a high-frequency camera, a vibration sensor, a rolling element cleaning mechanism, a rolling element detection controller, and a rolling element vibration detector. The rolling elements and the high-speed moving camera are mounted on the linear motion body. The rolling element image detection controller drives the rolling element tread cleaning mechanism. The high-frequency camera, vibration sensor, and rolling element cleaning mechanism are respectively installed at preset fixed positions on the track. The rolling element detection controller monitors and judges abnormal states of the rolling elements. The rolling element vibration detector receives vibration data. This invention enables real-time detection of defects in rolling elements under operating conditions, effectively solving the problem of difficult detection of rolling element defects during operation, and further ensuring the normal operation of the equipment.

Description

Technical Field

[0001] This invention relates to the field of material conveying technology, and more specifically to a device and method for detecting defects in rolling elements under operating conditions. Background Technology

[0002] In modern industrial production, defect detection of individual parts (assemblies) is a crucial step in ensuring the quality of finished products. However, the lack of effective online detection devices for rolling element components operating at high speeds presents a pressing technical challenge. These components play a vital role in critical equipment such as baggage handling, and their performance directly impacts the stability and safety of the entire system.

[0003] Currently, many companies rely on experience-based judgment when maintaining rolling element components, adopting preventative or reactive replacement strategies. While this approach can reduce the risk of failure to some extent, the lack of accurate data often leads to premature or delayed replacement decisions. Premature replacement not only increases user costs but may also waste resources; while delayed replacement may result in components failing to be replaced before wear and failure, potentially causing widespread equipment failure and greater losses.

[0004] Furthermore, the issue of quality consistency in rolling element components is a significant pain point. Due to deviations during the production process, different batches of components may exhibit significant performance differences. This inconsistency not only increases the overall operational instability of the equipment but can also lead to accidental damage, ultimately causing damage to other components and resulting in substantial economic losses.

[0005] Therefore, how to provide an advanced online detection device to detect the operating status and health status of rolling element components in real time is a problem that urgently needs to be solved by those skilled in the art. Summary of the Invention

[0006] In view of this, the present invention provides a detection device and method for rolling element defects under operating conditions, which can realize real-time detection of defects (such as wear and cracks) of rolling elements in the working state, effectively solving the problem of difficulty in monitoring rolling element defects during operation, thereby further ensuring the normal operation of equipment and improving overall reliability. At the same time, the present invention also addresses the wear problem of rolling elements after long-term operation, significantly reducing the impact risk and noise risk between worn rolling elements and the operating surface, thereby extending the service life of the rolling elements and reducing the operating cost of the equipment; it not only provides data support for maintenance strategies, helping enterprises to replace parts at the appropriate time, thereby reducing costs and improving equipment operating efficiency, but also effectively extends the service life of the equipment and reduces the chain reaction caused by component failure.

[0007] To achieve the above objectives, the present invention adopts the following technical solution:

[0008] A device for detecting rolling element defects under operating conditions includes a track, rolling elements, a high-speed moving camera, a linear motion body, a rolling element tread cleaning mechanism, a rolling element image detection controller, a high-frequency camera, a vibration sensor, a rolling element cleaning mechanism, a rolling element detection controller, and a rolling element vibration detector.

[0009] The rolling body and the high-speed moving camera are mounted on the linear motion body. The rolling body is rotatably connected to the track surface of the track, and the camera end of the high-speed moving camera corresponds to the plane of the rolling track surface of the rolling body.

[0010] The rolling element running tread cleaning mechanism is installed on the lower end face of the linear motion body; the rolling element image detection controller is configured to receive image information of the rolling track surface collected by the high-speed moving camera, monitor and judge the abnormal state of the rolling track surface through image processing and YOLO neural network algorithm, and drive the rolling element running tread cleaning mechanism to run according to whether there are foreign objects on the rolling track surface.

[0011] The high-frequency camera, the vibration sensor, and the rolling element cleaning mechanism are respectively installed at preset fixed positions on the track; the rolling element detection controller is configured to collect and monitor the image data and running status data of the rolling element in real time, and monitor and judge the abnormal state of the rolling element through image processing and YOLO neural network algorithm; the rolling element vibration detector is configured to receive the vibration data of the vibration sensor; the rolling element cleaning mechanism receives the data from the rolling element detection controller or the rolling element vibration detector to be controlled by it.

[0012] The beneficial effects of the above technical solution are as follows: the high-speed camera moves rapidly with the linearly moving body; the rolling element image detection controller and the high-speed camera monitor the image information of the track in real time through data interaction; the rolling element running tread cleaning mechanism cleans the track to ensure the cleanliness of the rolling track surface of the rolling element, thereby improving the detection accuracy of the rolling element; the high-frequency camera monitors the running status of the linearly moving body in real time; the rolling element detection controller ensures the transmission of high-frequency camera data and monitors and judges the rolling element status; the vibration sensor monitors the vibration information of the linearly moving body; the rolling element vibration detector collects and processes the vibration information; and the rolling element cleaning mechanism can clean foreign objects on the rolling element.

[0013] Preferably, there are multiple linear moving bodies, which are connected end-to-end by a chain structure. This chain structure forms a high-speed rolling element conveying system.

[0014] Preferably, the rolling element includes a guide rolling element and a traveling rolling element, with the rolling planes of the guide rolling element and the traveling rolling element arranged perpendicularly. The high-speed moving cameras are arranged in groups and respectively correspond to the rolling track planes of the guide rolling element and the traveling rolling element. The rolling of the guide rolling element and the traveling rolling element enables the linear motion body to be guided along the track and move at high speed.

[0015] Preferably, the high-frequency camera is located above the guide rolling element and the traveling rolling element, or multiple high-frequency cameras are fixed on the guide rolling element and the traveling rolling element respectively to simultaneously detect the operating status of the linear motion body, the guide rolling element, and the traveling rolling element. By monitoring the status of the linear motion body and the rolling element through the high-frequency cameras on the track and the rolling elements, defects in the operating state of the rolling elements can be monitored in real time.

[0016] Preferably, the rolling element detection controller is located at the terminal of the multiple high-frequency cameras to receive and process image data acquired by the multiple high-frequency cameras. The rolling element detection controller performs a series of image processing operations sequentially on the image data acquired by the high-frequency cameras, including image processing, error calculation, contour visualization, and neural network defect detection. The image processing stage covers background removal, grayscale processing, filtering and noise reduction, edge detection, contour tracking, contour filtering, and outer circle fitting, aiming to obtain optimized rolling element contour information. Subsequently, error analysis and visualization are performed on the contour information. Through systematic collection and training of various defect data of existing rollers, a precise neural network defect detection algorithm is further refined to achieve real-time monitoring and judgment of abnormal roller states.

[0017] Preferably, the high-frequency camera and the vibration sensor are arranged alternately along the length of the track. When the linearly moving body reaches the workstation equipped with the vibration sensor, the vibration sensor will collect raw vibration data in real time. To improve detection accuracy, the number of vibration sensors can be appropriately increased at different locations to achieve more precise monitoring.

[0018] Preferably, the rolling element cleaning mechanism can be attached to the linear motion body and communicate with the rolling element detection controller and the rolling element vibration detector to clean the rolling elements. The rolling element cleaning mechanism can respond to cleaning commands at any time and efficiently complete the cleaning task of the rolling elements.

[0019] The present invention also provides a method for detecting rolling element defects under operating conditions. The method uses the detection device in the above technical solution and employs the height method to detect rolling element defects. The principle is that the linear motion body moves around the track at high speed and the high-speed moving camera collects image data. The rolling element image detection controller displays the track surface information. When the track surface image is abnormal, the rolling element running tread cleaning mechanism cleans the track.

[0020] A high-frequency camera detects the contour of the rolling element and transmits the image data to the rolling element detection controller. The detection height of the rolling element is checked against the standard height. If the detection height is lower than the standard height, the rolling element is judged as defective. If the detection height is lower than the minimum limit of the standard height, the rolling element is judged as faulty. If the detection height is higher than the maximum limit of the standard height, the rolling element cleaning mechanism is activated to clean foreign objects from the rolling element.

[0021] The vibration sensor receives the vibration information of the rolling element and extracts the vibration information through the rolling element vibration detector. When the vibration information is greater than the normal range, the rolling element is judged to be abnormal. When the vibration information exceeds the limit value, the rolling element is judged to be faulty. When the vibration information is less than the normal range, the rolling element cleaning mechanism is activated and cleans the foreign objects on the rolling element.

[0022] The present invention also provides a method for detecting rolling element defects under operating conditions. The method uses the detection device in the above technical solution and adopts the circle center method to detect rolling element defects. The principle is that the linear motion body moves around the track at high speed and the high-speed moving camera collects image data. The rolling element image detection controller displays the track surface information. When the track surface image is abnormal, the rolling element running tread cleaning mechanism cleans the track.

[0023] A high-frequency camera detects the profile of the rolling element and transmits the image data to the rolling element detection controller. The detection center point of the rolling element is compared with the theoretical center point. If the detection center point is lower than the theoretical center point, the rolling element is determined to be defective. If the detection center point is lower than the minimum limit of the theoretical center point, the rolling element is determined to be faulty. If the detection center point is higher than the limit of the theoretical center point, the rolling element cleaning mechanism is activated and foreign objects on the rolling element are cleaned.

[0024] As can be seen from the above technical solution, compared with the prior art, the present invention discloses a detection device and method for detecting rolling element defects under operating conditions. This method enables real-time detection of defects (such as wear and cracks) in the rolling elements during operation, effectively solving the problem of difficult detection of rolling element defects during operation, thereby further ensuring the normal operation of the equipment and improving overall reliability. Simultaneously, the present invention also addresses the wear problem of rolling elements after prolonged operation, significantly reducing the impact risk and noise risk between worn rolling elements and the operating surface, thereby extending the service life of the rolling elements and reducing equipment operating costs. It not only provides data support for maintenance strategies, helping companies replace parts at appropriate times to reduce costs and improve equipment operating efficiency, but also effectively extends the service life of the equipment and reduces the chain reaction caused by component failure. Attached Figure Description

[0025] 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 only embodiments of the present invention. For those skilled in the art, other drawings can be obtained based on the provided drawings without creative effort.

[0026] Figure 1 A sectional view of a linearly moving body provided by the present invention;

[0027] Figure 2 This is a front view of the detection device provided by the present invention;

[0028] Figure 3 A top view of a linearly moving body provided by the present invention;

[0029] Figure 4 This is a schematic diagram of the rolling element height detection method provided by the present invention;

[0030] Figure 5 The flowchart of the intelligent rolling element detection algorithm provided by the present invention;

[0031] Figure 6 The flowchart of the rolling element vibration detection algorithm provided by the present invention is shown below;

[0032] Figure 7 This is a schematic diagram of the rolling element center detection method provided by the present invention;

[0033] Figure 8 This is a schematic diagram of the centerline of the rolling element provided by the present invention;

[0034] Figure 9 The flowchart of the rolling element defect height detection method provided by the present invention;

[0035] Figure 10 The flowchart of the rolling element defect center method detection provided by the present invention is shown.

[0036] in,

[0037] 1-Linear motion body; 2-High-speed moving camera; 3-Railway; 4-Rolling element; 5-Connector head; 6-Connector tail; 7-Rolling element running tread cleaning mechanism; 8-High frequency camera; 9-Rolling element cleaning mechanism; 10-Vibration sensor; 11-Rolling element image detection controller; 12-Rolling element detection controller; 13-Rolling element vibration detector. Detailed Implementation

[0038] 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, and 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.

[0039] Example 1:

[0040] See appendix Figures 1-3 This invention discloses a device for detecting defects in rolling elements under operating conditions, comprising a track 3, rolling elements 4, a high-speed moving camera 2, a linear motion body 1, a rolling element tread cleaning mechanism 7, a rolling element image detection controller 11, a high-frequency camera 8, a vibration sensor 10, a rolling element cleaning mechanism 9, a rolling element detection controller 12, and a rolling element vibration detector 13.

[0041] The rolling body 4 and the high-speed moving camera 2 are mounted on the linear motion body 1. The rolling body 4 is tumblingly connected to the track surface of the track 3, and the camera end of the high-speed moving camera 2 corresponds to the plane of the rolling track surface of the rolling body 4.

[0042] The rolling element running tread cleaning mechanism 7 is installed on the lower end face of the linear motion body 1; the rolling element image detection controller 11 is configured to receive image information of the rolling track surface collected by the high-speed moving camera 2, monitor and judge the abnormal state of the rolling track surface through image processing and YOLO neural network algorithm, and drive the rolling element running tread cleaning mechanism 7 to run according to whether there are foreign objects on the rolling track surface.

[0043] The high-frequency camera 8, vibration sensor 10, and rolling element cleaning mechanism 9 are respectively installed at preset fixed positions on the track 3; the rolling element detection controller 12 is configured to collect and monitor the image data and running status data of the rolling element 4 in real time, and monitor and judge the abnormal state of the rolling element 4 through image processing and YOLO neural network algorithm; the rolling element vibration detector 13 is configured to receive the vibration data of the vibration sensor 10; the rolling element cleaning mechanism 9 receives the data from the rolling element detection controller 12 or the rolling element vibration detector 13 to be controlled by it.

[0044] like Figure 1 As shown, the track has two mutually perpendicular track surfaces. The rolling body 4 includes a guide rolling body and a traveling rolling body. The rolling planes of the guide rolling body and the traveling rolling body are arranged perpendicularly. The high-speed moving camera 2 is arranged in groups and is respectively on the plane of the rolling track surface of the guide rolling body and the traveling rolling body.

[0045] High-speed cameras can accurately and in real time acquire image information of mutually perpendicular track surfaces and present a complete track surface projection in the rolling element image detection controller. The algorithm on the rolling element image detection controller deeply analyzes and calculates the state of the track surface and works in conjunction with the rolling element running tread cleaning mechanism to successfully complete the track surface cleaning task, ensuring that the track surface is free of foreign object interference and guaranteeing the normal operation of the linear motion body.

[0046] To further optimize the above technical solution, the number of linear motion bodies 1 is multiple, and the multiple linear motion bodies 1 are connected end-to-end using a chain structure. For example... Figure 3 As shown, one end of the linear motion body 1 is fixed with a connecting head 5 and the other end is fixed with a connecting tail 6. Both the connecting head 5 and the connecting tail 6 are provided with standard pitch connecting holes. Two adjacent linear motion bodies 1 are connected by a chain structure. The guide rolling body is located on the side wall of the linear motion body 1, and the traveling rolling body is located at the bottom of the linear motion body 1 and the two are arranged perpendicular to each other.

[0047] To further optimize the above technical solution, the high-frequency camera 8 is located above the guide rolling body and the traveling rolling body, or multiple high-frequency cameras 8 are fixed on the guide rolling body and the traveling rolling body respectively to simultaneously detect the running status of the linear motion body 1 and the guide rolling body and the traveling rolling body.

[0048] See appendix Figure 4The guide rolling body and the traveling rolling body are each equipped with four high-frequency cameras, and the angles of these four high-frequency cameras relative to the guide rolling body or the traveling rolling body are 90°. According to the requirements of detection accuracy, the number of linear moving bodies carrying high-frequency cameras can be adjusted to achieve a higher detection effect.

[0049] To further optimize the above technical solution, the rolling element detection controller 12 is located at the terminal of multiple high-frequency cameras 8 to receive and process the image data collected by the multiple high-frequency cameras 8.

[0050] like Figure 5 As shown, the rolling element detection controller is located at the terminal of the high-frequency camera. It performs a series of image processing operations on the image data acquired by the high-frequency camera, including image processing, error calculation, contour visualization, and neural network defect detection. The image processing stage uses background subtraction to remove the background, OpenCV color conversion, Gaussian filtering for noise reduction, Canny edge detection, OpenCV outer circle contour search, contour filtering, and outer circle fitting to obtain optimized rolling element contour information. Subsequently, error analysis and visualization are performed on the contour information. Through systematic collection and training of various existing rolling element defect data (500 training images, 100 defect images, and 50 validation images), a precise YOLO neural network defect detection algorithm is further refined to achieve real-time monitoring and judgment of abnormal rolling element states.

[0051] To further optimize the above technical solutions, the YOLO neural network defect detection algorithm not only covers the automatic identification and timely judgment of rolling element contours, but also accurately identifies various defects in rolling elements, such as cracks, dents, and damage. The dataset trained on the YOLO neural network defect detection algorithm originates from real-time extraction from a high-frequency camera, which can effectively improve the accuracy of detection.

[0052] The algorithm achieved an accuracy of 98% on the validation set, and no misclassifications were found in subsequent experiments. Furthermore, for any misclassified results, staff can manually modify the labels to further optimize detection accuracy.

[0053] To further optimize the above technical solution, the high-frequency camera 8 and the vibration sensor 10 are arranged alternately along the length of the track 3. When the linearly moving body reaches the station equipped with the vibration sensor, the vibration sensor will collect raw vibration data in real time. To improve detection accuracy, the number of vibration sensors can be appropriately increased at different locations to achieve more precise monitoring.

[0054] like Figure 6As shown, the rolling element vibration detector receives vibration data collected by the vibration sensor in real time and sequentially performs a series of operations such as data acquisition, filtering, data standardization, time domain analysis, frequency domain analysis, anomaly detection, fault mode identification, decision model construction, and comprehensive evaluation to accurately obtain the vibration information of each component and effectively determine its fault status.

[0055] To further optimize the above technical solution, the rolling element cleaning mechanism 9 can be attached to the linear motion body 1 and communicate with the rolling element detection controller 12 and the rolling element vibration detector 13 to clean the rolling elements 4. The rolling element cleaning mechanism can respond to cleaning commands at any time and efficiently complete the cleaning task of the rolling elements.

[0056] Example 2:

[0057] like Figures 1-6 and Figure 9 As shown, this embodiment of the invention discloses a method for detecting rolling element defects under operating conditions. The method uses the rolling element defect detection device under operating conditions described in Embodiment 1 to detect rolling element defects, employing the height method.

[0058] This embodiment takes two sets of identical structures as examples, namely two sets of linear motion bodies carrying rolling tread cleaning mechanisms and two sets of linear motion bodies carrying high-speed moving cameras, two sets of high-frequency cameras fixed at preset track positions, and two sets of vibration sensors fixed at preset track positions.

[0059] In this embodiment, the rolling element image detection controller 11 is located at the terminal of the high-speed moving camera 2; the rolling element detection controller 12 is located at the terminal of the high-frequency camera 8; and the rolling element vibration detector 13 is located at the terminal of the vibration sensor 10. All three are integrated into one terminal.

[0060] Its detection principle is:

[0061] The linear motion body 1 moves at high speed around the track 3 and collects image data through the high-speed moving camera 2. The rolling body image detection controller 11 displays the track surface information of the track 3. When the track surface image is abnormal (protrusion, depression), the linear motion body carrying the rolling body running tread cleaning mechanism 7 cleans the track 3 until the rolling body image detection controller 11 detects a normal track.

[0062] The high-frequency camera 8 on track 3 detects the contour of the rolling body 4 and transmits the image data to the rolling body detection controller 12. The algorithm mounted on the rolling body detection controller 12 then finds the highest point of the rolling body 4. Figure 4 (8A, 8D, 8G, 8E) and the lowest point ( Figure 4 (8B, 8C, 8H, 8F) , automatically calculate the height difference between two opposing high-frequency cameras ( Figure 4The system uses H1, H2, H3, and H4 in the diagram to check the detected height of the rolling element 4 against the standard height (H1, H2, H3, and H4 are compared with H in the diagram). When the detected height is lower than the standard height, the rolling element 4 is determined to be defective, but it can continue to be used within a certain range and its height information is marked. When the detected height is lower than the minimum limit of the standard height, the rolling element 4 is determined to be faulty. In particular, when the same rolling element 4 has three different heights higher than the marked height information (lower than the standard range, but not lower than the minimum limit height of the rolling element 4), the rolling element 4 is determined to be faulty. When the detected height is higher than the maximum limit of the standard height, the rolling element cleaning mechanism 9 is activated and cleans the foreign objects on the rolling element 4.

[0063] Meanwhile, the vibration sensor 10 receives the vibration information of the rolling element 4 and extracts the vibration information through the rolling element vibration detector 13. When the detected vibration information is greater than the normal range, the rolling element 4 is determined to be abnormal and can continue to be used within a certain range. When the vibration information exceeds the limit range, the rolling element 4 is determined to be faulty. In particular, when the same rolling element 4 detects abnormal vibration information more than five times with different situations, the rolling element 4 is determined to be faulty. When the vibration information is less than the normal range and there is slight abnormal information, the rolling element cleaning mechanism 9 is activated and cleans the foreign objects on the rolling element 4.

[0064] To further optimize the above technical solution, if the same rolling element 4 is judged as both defective and abnormal, then the rolling element 4 will be re-judged as faulty.

[0065] The detection accuracy in this embodiment is closely related to the number and layout of high-frequency cameras and vibration sensors in the entire system. Regarding the number of cameras, it is recommended to use an even number to reduce the randomness of the system. Regarding the layout, high-frequency cameras should be rationally configured along the circumference of the rolling element to minimize duplicate detection points.

[0066] Example 3:

[0067] See appendix Figures 1-5 and Figure 7 , Figure 8 , Figure 10 This invention discloses a method for detecting rolling element defects under operating conditions. The method uses the rolling element defect detection device under operating conditions described in Embodiment 1 to detect rolling element defects, employing the center-of-circle method.

[0068] In this embodiment, the rolling element image detection controller 11 is located at the terminal of the high-speed moving camera 2; the rolling element detection controller 12 is located at the terminal of the high-frequency camera 8; and the rolling element vibration detector 13 is located at the terminal of the vibration sensor 10. All three are integrated into one terminal.

[0069] This embodiment takes two sets of identical structures as examples, namely two sets of linear motion bodies carrying rolling tread cleaning mechanisms and two sets of linear motion bodies carrying high-speed moving cameras, two sets of high-frequency cameras fixed at preset track positions, and two sets of vibration sensors fixed at preset track positions.

[0070] Its detection principle is:

[0071] The linear motion body 1 moves at high speed around the track 3 and collects image data through the high-speed moving camera 2. The rolling body image detection controller 11 displays the track surface information of the track 3. When the track surface image is abnormal (protrusion, depression), the linear motion body carrying the rolling body running tread cleaning mechanism 7 cleans the track 3 until the rolling body image detection controller 11 detects a normal track.

[0072] The high-frequency camera 8 detects the contour of the rolling element 4 and transmits the image data to the rolling element detection controller 12. The algorithm mounted on the rolling element detection controller 12 then finds the highest point of the rolling element 4. Figure 7 (8A, 8D, 8G, 8E) and the lowest point ( Figure 7 (8B, 8C, 8H, 8F) The detection center point of rolling body 4 is checked against the theoretical center point, and the center point of the highest and lowest points between the two opposing high-frequency cameras 8 on rolling body 4 is automatically calculated. Figure 7 R1, R2, R3, R4, R5, R6 in the equation, and the theoretical center point of rolling element 4 ( Figure 7 The rolling element 4 is compared with the theoretical center point (R) in the standard range (R-Rmin). When the detected center point (R3, R4) is lower than the theoretical center point (R), the rolling element 4 is determined to be defective. It is allowed to continue to be used within a certain range (R~Rmin) and the position information of the center point (R3 and R4) of the rolling element is marked. When the detected center point (R5, R6) is lower than the minimum limit value (Rmin) of the theoretical center point, the rolling element 4 is determined to be faulty. In particular, when the center point of the same rolling element 4 is higher than the marked position information three times (lower than the standard range (R~Rmin) and not lower than the minimum limit height (Rmin) of the rolling element), the rolling element 4 is determined to be faulty. When the detected center point (R7, R8) is higher than the limit value (R~Rmax) of the theoretical center point, the rolling element cleaning mechanism 9 is activated and cleans the foreign objects on the rolling element 4.

[0073] To further optimize the above technical solutions, such as Figure 8 As shown, by detecting the center point (R1, R2, R3, R4, R5, R6, R7, R8) of each rolling element, the actual center point trajectory of each rolling element can be generated. By combining the trajectory image, the theoretical center point image (R01) of the rolling element, and the actual center point (R) trajectory of the rolling element, the running status of each rolling element can be obtained. If a rolling element has a clear defect trend, further monitoring and control can be implemented.

[0074] The detection accuracy in this embodiment is closely related to the number and layout of high-frequency cameras and vibration sensors in the entire system. Regarding the number of cameras, it is recommended to use an even number to reduce the randomness of the system. Regarding the layout, high-frequency cameras should be rationally configured along the circumference of the rolling element to minimize duplicate detection points.

[0075] The various embodiments in this specification are described in a progressive manner, with each embodiment focusing on its differences from other embodiments. Similar or identical parts between embodiments can be referred to interchangeably. For the apparatus disclosed in the embodiments, since they correspond to the methods disclosed in the embodiments, the description is relatively simple; relevant parts can be referred to the method section.

[0076] The above description of the disclosed embodiments enables those skilled in the art to make or use the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features disclosed herein.

Claims

1. A device for detecting a defect of a rolling element in a running condition, characterized in that, The track (3), the rolling body (4), the high-speed moving camera (2), the linear motion body (1), the rolling body running tread cleaning mechanism (7), the rolling body image detection controller (11), the high-frequency camera (8), the vibration sensor (10), the rolling body cleaning mechanism (9), the rolling body detection controller (12) and the rolling body vibration detector (13), The rolling body (4) and the high-speed moving camera (2) are installed on the linear motion body (1), the rolling body (4) is connected on the track surface of the track (3), and the imaging end of the high-speed moving camera (2) corresponds to the rolling track surface of the rolling body (4). The number of the linear motion body (1) is multiple, and the multiple linear motion bodies (1) are connected in a chain structure; the rolling body (4) includes a guide rolling body and a walking rolling body, the rolling planes of the guide rolling body and the walking rolling body are arranged vertically, and the high-speed moving cameras (2) are arranged in groups and correspond to the rolling track surfaces of the guide rolling body and the walking rolling body respectively. The rolling body running tread cleaning mechanism (7) is installed on the lower end surface of the linear motion body (1); the rolling body image detection controller (11) is configured to receive image information of the rolling track surface collected by the high-speed moving camera (2), monitor and judge the abnormal state of the rolling track surface through image processing and YOLO neural network algorithm, and drive the rolling body running tread cleaning mechanism (7) to operate according to whether there is foreign matter on the rolling track surface. The high-frequency camera (8), the vibration sensor (10) and the rolling body cleaning mechanism (9) are respectively installed at the preset fixed positions on the track (3); the rolling body detection controller (12) is configured to collect and monitor the image data and the running state data of the rolling body (4) in real time, monitor and judge the abnormal state of the rolling body (4) through image processing and YOLO neural network algorithm; the rolling body vibration detector (13) is configured to receive the vibration data of the vibration sensor (10); and the rolling body cleaning mechanism (9) receives the data of the rolling body detection controller (12) or the rolling body vibration detector (13) to be controlled and operated.

2. The device for detecting a defect of a rolling element in an operating condition according to claim 1, characterized in that, The high-frequency camera (8) is located above the guide rolling body and the walking rolling body, or multiple high-frequency cameras (8) are respectively fixed on the guide rolling body and the walking rolling body to simultaneously detect the running state of the linear motion body (1), the guide rolling body and the walking rolling body.

3. The device for detecting a defect of a rolling element in an operating condition according to claim 2, characterized in that, The rolling body detection controller (12) is located at the terminal of the multiple high-frequency cameras (8) to receive and process the image data collected by the multiple high-frequency cameras (8).

4. The device for detecting defects of rolling elements in operation according to claim 1, characterized in that, The high-frequency camera (8) and the vibration sensor (10) are staggered along the length direction of the track (3).

5. A method for detecting defects of rolling elements under operating conditions, using the device for detecting defects of rolling elements under operating conditions according to any one of claims 1 to 4, characterized in that, The linear motion body (1) moves at high speed around the track (3) and collects image data through the high-speed moving camera (2), the rolling body image detection controller (11) displays the track surface information of the track (3), when the track surface image is abnormal, the rolling body running tread cleaning mechanism (7) cleans the track (3); The high-frequency camera (8) detects the profile of the rolling body (4) and transmits the image data to the rolling body detection controller (12), and checks the detection height of the rolling body (4) with the standard height, when the detection height is lower than the standard height, the rolling body (4) is judged as defective; when the detection height is lower than the lowest limit value of the standard height, the rolling body (4) is judged as a fault; when the detection height is higher than the highest limit value of the standard height, the rolling body cleaning mechanism (9) starts and cleans the foreign matter on the rolling body (4); The vibration sensor (10) receives the vibration information of the rolling body (4) and extracts the vibration information through the rolling body vibration detector (13), when the vibration information is greater than the normal range, the rolling body (4) is judged as abnormal, when the vibration information exceeds the limit value, the rolling body (4) is judged as a fault, when the vibration information is less than the normal range, the rolling body cleaning mechanism (9) starts and cleans the foreign matter on the rolling body (4).

Images