A device for detecting the addendum flatness of a gear

By combining a reflector and an image sensor, efficient and accurate detection of gear tooth tips is achieved, solving the problem of high workload in tooth tip detection in existing technologies and improving detection efficiency and accuracy.

CN120668062BActive Publication Date: 2025-11-07CHONGQING GEARBOX & MASCH CO LTD
View PDF 3 Cites 0 Cited by

Patent Information

Application Number
CN202511179516.1
Authority / Receiving Office
CN · China
Patent Type
Patents(China)
Current Assignee / Owner
Filing Date
2025-08-22
Publication Date
2025-11-07
Estimated Expiration
2045-08-22

AI Technical Summary

Technical Problem

Existing gear tooth tip inspection equipment requires inspection of each tooth tip separately, which is labor-intensive and difficult to complete efficiently.

Method used

A reflector is arranged around the gear, and an image sensor is used to collect reflected images of multiple tooth tips. A convolutional neural network is used for image recognition, and combined with a laser beam switch and automatic alignment function, unified detection of multiple tooth tips is achieved.

Benefits of technology

It reduces the workload of repeated inspection of each tooth tip, improves inspection efficiency, and ensures the accuracy and consistency of inspection through image recognition and automatic alignment functions.

✦ Generated by Eureka AI based on patent content.

Smart Images

  • Figure CN120668062B_ABST
    Figure CN120668062B_ABST
Patent Text Reader

Abstract

The present application relates to the field of mechanical detection, and particularly relates to a tooth crest flatness detection device for gear, which comprises a containing table and a controller, the containing table is provided with a matching column, the matching column is used for installing the gear to be detected; the containing table is provided with a reflecting frame, the reflecting frame surrounds the matching column, the inner side of the reflecting frame is provided with a plurality of inclined mirrors; the top of the reflecting frame is provided with a cross frame, the bottom of the cross frame is fixedly connected with an image sensor, the image sensor is used for collecting the images reflected by the mirrors; the controller is used for acquiring the images collected by the image sensor, and judging whether the tooth crest of the gear is flat based on image recognition.
Need to check novelty before this filing date? Find Prior Art

Description

TECHNICAL FIELD

[0001] The present application relates to the field of mechanical detection, in particular to a tooth tip flatness detection device for gear. BACKGROUND

[0002] The tooth tip is the tooth surface contained by the tooth tip surface, which has a clear positioning feature in the gear geometry. For external gear, the tooth tip surface is the surface farthest from the axis of the tooth body, while for internal gear, it is the surface closest to the axis. The tooth tip surface is specialized as a tooth tip cylindrical surface in cylindrical gear, and the boundary size is determined by the tooth tip circle diameter formula. During meshing, the tooth tip surface participates in the formation of the effective tooth surface contact area, directly affecting the sliding coefficient distribution and wear characteristics of the gear pair. In engineering applications, tooth tip modification techniques are often used to optimize contact stress distribution and avoid edge wear.

[0003] In the prior art, for example, CN119334230B discloses a tooth tip flatness detection device for gear, which improves the accuracy of detection by removing the tooth tip dust of the gear. However, the gear has many teeth, resulting in each tooth tip being independent. The conventional detection method uses a measuring tool to measure each tooth tip, which requires a large amount of work. SUMMARY

[0004] To solve the above problems, the present application provides a tooth tip flatness detection device for gear, which reduces the workload of tooth tip detection through a mirror.

[0005] To achieve the above purpose, the technical scheme of the present application is as follows: a tooth tip flatness detection device for gear, comprising a holding table and a controller, the holding table is provided with a matching column, the matching column is used to install the gear to be detected;

[0006] The holding table is provided with a reflection frame, the reflection frame surrounds the matching column, and the inner side of the reflection frame is provided with a plurality of inclined mirrors;

[0007] The top of the reflection frame is provided with a horizontal frame, and the bottom of the horizontal frame is fixedly connected with an image sensor, the image sensor is used to collect the images reflected by each mirror;

[0008] The controller is used to obtain the images collected by the image sensor, and to judge whether the tooth tip of the gear is flat based on image recognition.

[0009] The above scheme has the following beneficial effects:

[0010] 1. In this solution, when the user performs gear tooth tip flatness inspection, the gear is coaxially fixed to the mating column. Reflectors are arranged around the gear, ensuring that the image of all gear tooth tips is reflected. The image sensor, located at the bottom of the crossbeam, can capture the reflected images of the tooth tips. This unifies the images of multiple tooth tips into a single viewpoint, allowing the image sensor to analyze the condition of multiple tooth tips in a single acquisition. Compared to existing technologies, the design of the reflector frame and mirrors eliminates the need for the user to inspect each tooth tip separately; instead, the reflectors reflect each tooth tip of the gear into the image sensor's imaging range for inspection.

[0011] 2. In this solution, the controller identifies whether there is unevenness on the tooth tip through image recognition, marks the uneven areas on the tooth tip, and completes the flatness detection.

[0012] Furthermore, the reflector has a tilt angle of 45 degrees, and the reflector is detachably connected to the inner side of the reflector mount.

[0013] Beneficial effect: The reflector is tilted at 45 degrees, which reflects horizontal light into vertical light so that the image sensor placed at the bottom of the crossbeam can capture it.

[0014] Furthermore, the inner side of the reflector frame is provided with a hook and loop fastener, and the back of the reflector is provided with a hook and loop fastener. The reflector and the reflector frame are detachably connected through the hook and loop fastener and the hook and loop fastener.

[0015] Beneficial effects: Different gears have different numbers of teeth and diameters. To accommodate the testing of different gears, the reflector is designed to be detachably connected to the reflector frame via Velcro hooks and loops. This allows users to adapt the reflector to the gear being tested by installing different reflectors.

[0016] Furthermore, the width of the imaging range of the gear tooth tip of the reflector corresponds to the width of the gear tooth tip.

[0017] Beneficial effect: An excessively wide reflector may project the tooth root or tooth surface into the reflected image, causing the reflector to reflect other areas of the gear. This necessitates the controller identifying the position of the tooth tip in the acquired image. However, the reflector's tooth tip imaging range corresponds to the tooth tip width, ensuring that only the tooth tip is visible in the image, eliminating the need to differentiate between the tooth tip and other parts of the gear.

[0018] Furthermore, the display platform and the crossbeam are equipped with paired laser beam switches, the monitoring range of which is aligned with the edge of the imaging range of the reflector.

[0019] Beneficial effect: when detecting, the gear tooth top and the mirror need to be aligned, the laser emitting and receiving switch can be triggered after the gear tooth touches the laser emitting and receiving switch, which helps the user to determine the position of the gear tooth and align the gear tooth top and the mirror.

[0020] Further, the matching column is provided with a motor for driving the matching column to rotate, and the controller is used to control the motor to drive the matching column to rotate, and the laser emitting and receiving switch is used to judge whether the gear tooth reaches the monitoring range of the laser emitting and receiving switch, and if so, the controller controls the motor to brake.

[0021] Beneficial effect: the matching column can be driven to rotate by the motor, so that after the gear tooth reaches the monitoring range of the laser emitting and receiving switch, the motor is automatically controlled to brake, and the automatic alignment of the gear is completed.

[0022] Further, the bottom of the cross frame is provided with a telescopic rod, and the end of the telescopic rod away from the cross frame is fixedly connected with a blowing cylinder.

[0023] Beneficial effect: the telescopic rod can be extended before detection to blow dust on the gear tooth top through the blowing cylinder, so as to reduce the influence of dust attached to the gear tooth top on subsequent detection.

[0024] Further, the bottom of the cross frame is provided with an annular slide rail, and the telescopic rod is slidably connected to the slide rail, and the slide rail is provided with a driving member for driving the telescopic rod to slide along the slide rail.

[0025] Beneficial effect: the driving member can drive the telescopic rod to move along the slide rail, so as to move around the gear and clean the dust on the gear in a circular manner.

[0026] Further, the end of the telescopic rod away from the cross frame is also fixedly connected with a wiping block, and the wiping block is located at the back of the blowing cylinder, and the wiping block is used to clean the mirror when the driving member drives the telescopic rod to slide.

[0027] Beneficial effect: the cleanliness of the mirror also affects the imaging process, and the wiping block is located at the back of the blowing cylinder, so that when the blowing cylinder cleans the gear, the wiping block can also wipe the mirror to maintain the cleanliness of the mirror.

[0028] Further, the controller is provided with a trained convolutional neural network, and the convolutional neural network is trained based on a plurality of stepped protrusions, depressions, collapses, scratches and texture image samples of the gear tooth top, and the convolutional neural network is used to input the image collected by the image sensor and frame the stepped protrusions, depressions, collapses, scratches and textures in the image.

[0029] Beneficial effect: the controller is provided with a convolutional neural network, which is used to identify the image collected by the image sensor to analyze the flatness of the gear tooth top.

[0030] Additional aspects and advantages of the present application will be made apparent from the following description. BRIEF DESCRIPTION OF DRAWINGS

[0031] Figure 1 Axonometric view of the embodiment of the present application for detecting the addendum flatness of a gear;

[0032] Figure 2 Schematic view of the reflecting frame of the embodiment of the present application for detecting the addendum flatness of a gear;

[0033] Figure 3 Schematic view of the ring-shaped slide rail of the embodiment of the present application for detecting the addendum flatness of a gear;

[0034] Figure 4 Schematic view of the side view structure of the embodiment of the present application for detecting the addendum flatness of a gear;

[0035] Figure 5 Schematic view of the lifting structure and gear cooperation of the embodiment of the present application for detecting the addendum flatness of a gear;

[0036] Figure 6 Axonometric view of the lifting structure of the embodiment of the present application for detecting the addendum flatness of a gear;

[0037] Figure 7 Enlarged view of part A of the embodiment of the present application for detecting the addendum flatness of a gear;

[0038] Figure 8 Enlarged view of part B of the embodiment of the present application for detecting the addendum flatness of a gear.

[0039] The reference signs in the drawings of the specification include: 1, holding table; 2, cooperation column; 3, gear; 4, reflecting frame; 5, reflecting mirror; 6, horizontal frame; 7, image sensor; 8, laser transmission switch; 9, telescopic rod; 10, blowing cylinder; 11, ring-shaped slide rail; 12, wiping block; 13, lifting table; 14, detection ring; 15, through slot; 16, scraping block; 17, air bag strip. DETAILED DESCRIPTION

[0040] The technical solutions of the present application will be described clearly and completely below in conjunction with the drawings. Obviously, the described embodiments are part of the embodiments of the present application, rather than all the embodiments. Based on the embodiments in the present application, all other embodiments obtained by those skilled in the art without creative labor fall within the scope of protection of the present application.

[0041] In the description of this invention, it should be noted that the terms "center," "upper," "lower," "left," "right," "vertical," "horizontal," "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," "second," and "third" are used for descriptive purposes only and should not be construed as indicating or implying relative importance.

[0042] In the description of this invention, it should be noted that, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention based on the specific circumstances.

[0043] The following detailed description illustrates the specific implementation method:

[0044] Example 1:

[0045] As attached Figure 1 As shown, a gear tooth tip flatness testing device includes a holding platform 1 and a controller. A mating column 2 is bolted to the holding platform 1. The end of the mating column 2 is provided with a spline for mating with the gear 3 to be tested. The mating column 2 is used to install the gear 3 to be tested.

[0046] The display platform 1 is bolted to a reflector frame 4, which surrounds the mounting column 2. Several tilted reflectors 5 are detachably connected to the inner side of the reflector frame 4.

[0047] The top of the reflector frame 4 is equipped with a crossbeam 6, which is fixedly connected to the display platform 1 by a support column. Figure 3 As shown, an image sensor 7 is bolted to the bottom of the crossbeam 6. The image sensor 7 is a wide-angle camera used to capture images reflected by each reflector 5. The reflector 5 has a tilt angle of 45 degrees. The inner side of the reflector frame 4 has a Velcro surface, and the back of the reflector 5 has a Velcro hook surface. The reflector 5 and the reflector frame 4 are detachably connected via the Velcro surface and the Velcro hook surface. The number of reflectors 5 corresponds to the number of tooth tips of the gear 3. The imaging range width of the gear 3 tooth tip of the reflector 5 corresponds to the width of the gear 3 tooth tip.

[0048] The controller is internally provided with a trained convolutional neural network, the convolutional neural network is trained based on a plurality of stepped protrusions, depressions, collapses, scratches and texture image samples of the tooth top of the gear 3, the image samples are images reflected by the 40°-50° reflector 5, the number of image samples is greater than 1800, the learning rate of the convolutional neural network is between 0.01-0.05, and the convolutional neural network is used for inputting images collected by the image sensor 7 and framing the stepped protrusions, depressions, collapses, scratches and textures in the marked image. The controller is used to acquire the images collected by the image sensor 7 and judge whether the tooth top of the gear 3 is flat based on image recognition.

[0049] When the user detects the flatness of the tooth top of the gear 3, the gear 3 is coaxially fixed on the matching column 2. The reflector 5 is arranged around the gear 3, so that the tooth top picture of the gear 3 is reflected by the reflector 5. The image sensor 7 is located at the bottom of the cross frame 6 and can collect the reflected picture of the tooth top. Then, the pictures of a plurality of tooth tops are unified in a single view, so that the image sensor 7 can analyze the conditions of a plurality of tooth tops through single collection. Compared with the prior art, through the design of the reflection frame 4 and the reflector 5, the user does not need to detect each tooth top respectively, but uses the reflector 5 to reflect each tooth top of the gear 3 into the imaging range of the image sensor 7 for detection.

[0050] The inclination angle of the reflector 5 is 45 degrees, and the horizontal light is reflected as vertical light for the image sensor 7 placed at the bottom of the cross frame 6 to collect. Different gears 3 have different numbers of teeth and diameters. In order to adapt to the detection of different gears 3, the reflector 5 is arranged to be detachably connected with the reflection frame 4 through the magic tape rough surface and the magic tape hook surface, so that the user can adapt by installing different reflectors 5 according to the different gears 3 to be detected.

[0051] If the reflector 5 is too wide, the dedendum or tooth surface may be reflected into the reflected picture, so that the reflector 5 reflects other areas of the gear 3. This makes the controller need to identify the position of the tooth top in the collected image. The imaging range width of the gear 3 tooth top of the reflector 5 corresponds to the tooth top width of the gear 3, so that only the tooth top picture exists in the image, without distinguishing the tooth top and other parts of the gear 3, which can reduce the algorithmic pressure of image recognition.

[0052] Embodiment 2:

[0053] As shown in Figure 1 , Figure 2 and Figure 3 , the holding table 1 and the cross frame 6 are provided with a pair of laser beam receiving switches 8, and the monitoring range of the laser beam receiving switch 8 is aligned with the edge of the imaging range of the reflector 5. In addition, the specific installation position of the laser beam receiving switch 8 can be adaptively adjusted according to the actual situation.

[0054] The motor is used for driving the cooperation column 2 to rotate, and the controller is used for controlling the motor to drive the cooperation column 2 to drive the gear 3 to rotate. The laser emitting and receiving switch 8 is used for judging whether the gear tooth of the gear 3 reaches the monitoring range of the laser emitting and receiving switch 8. If the gear tooth reaches the monitoring range of the laser emitting and receiving switch 8, the controller controls the motor to brake.

[0055] The bottom of the horizontal frame 6 is bolted with a telescopic rod 9, the telescopic rod 9 is a cylinder, and the end of the telescopic rod 9 away from the horizontal frame 6 is bolted with a blowing cylinder 10, the blowing cylinder 10 is preferably a fan and is electrically connected with the controller. The bottom of the horizontal frame 6 is provided with an annular slide rail 11, the telescopic rod 9 is slidably connected to the slide rail, and the slide rail is provided with a driving member for driving the telescopic rod 9 to slide along the slide rail, the driving member is a roller. The end of the telescopic rod 9 away from the horizontal frame 6 is bolted with a wiping block 12, the wiping block 12 is located at the back of the blowing cylinder 10, the material of the wiping block 12 is sponge, and the inclination angle of the wiping block 12 is 45°. The wiping block 12 is used for cleaning the reflecting mirror 5 when the driving member drives the telescopic rod 9 to slide.

[0056] When detecting, the gear tooth top of the gear 3 needs to be aligned with the reflecting mirror 5. The laser emitting and receiving switch 8 can be triggered after the gear tooth of the gear tooth top touches the ray of the laser emitting and receiving switch 8, which helps the user to determine the position of the gear tooth and align the gear tooth top with the reflecting mirror 5. The cooperation column 2 can be driven to rotate by the motor, so that after the gear tooth of the gear 3 reaches the monitoring range of the laser emitting and receiving switch 8, the motor is automatically controlled to brake, and the automatic alignment of the gear 3 is completed.

[0057] The telescopic rod 9 can be extended before detection to blow dust on the gear tooth top of the gear 3 by the blowing cylinder 10, so as to reduce the influence of dust attached to the gear tooth top of the gear 3 on subsequent detection. The driving member can drive the telescopic rod 9 to move along the slide rail, so as to move around the gear 3 and clean the dust on the gear 3 in a surrounding manner. The cleanliness of the reflecting mirror 5 also affects the imaging process. The wiping block 12 is located at the back of the blowing cylinder 10. When the blowing cylinder 10 cleans the gear 3, the wiping block 12 can also wipe the reflecting mirror 5 to maintain the cleanliness of the reflecting mirror 5.

[0058] Embodiment 3:

[0059] Different from the above embodiments, in actual use, scratches on the surface of the reflecting mirror 5, light fluctuation during detection, oil stains and reflection on the surface (gear tooth top surface) of the gear 3, etc. may be misjudged as gear tooth defects (for example, a hair on the reflecting mirror 5 may be identified as a "scratch" in the image). The training samples of the convolutional neural network are difficult to cover all extreme working conditions (such as special angle concave reflection), and "false positive" (misjudging qualified as unqualified) or "false negative" (misjudging unqualified as qualified) may occur. Therefore, in order to further reduce the risk of misjudgment, the present scheme, for example, Figure 5As shown, the top end of the holding table 1 is screw-connected with a lifting table 13 electrically connected with the controller, the lifting table 13 is annular, and the output end of the lifting table 13 is screw-connected with a detection ring 14, and the detection ring 14 is combined with Figure 6 and Figure 7 As shown, a plurality of through grooves 15 corresponding to the teeth of the gear 3 are formed in the inner ring of the detection ring 14, and a plurality of air bag strips 17 are adhered to the inner side of the through grooves 15 from top to bottom. The air bag strips 17 are made of elastic material (such as silica gel) and are inflated to keep an expanded state, so that they can closely fit the curved surface shape of different tooth tips (whether the tooth tip is a plane or a micro-arc surface), avoiding damage to the tooth tip by rigid detection tools, while ensuring that the subsequent pressure signals can truly reflect the surface concave-convex changes. The scraping blocks 16 are welded on the inner side of the through grooves 15, and the scraping blocks 16 are located above the air bag strips 17, and the surface of the scraping blocks 16 is adhered with a sponge layer. The position of the scraping blocks 16 is close to the top end of the detection ring 14 and is arranged obliquely (such as 30° obliquely upward), so that the dust or impurities scraped can move along the inclined surface of the scraping blocks 16 to the acute angle between the scraping blocks 16 and the detection ring, thereby achieving the collection and limitation, facilitating subsequent cleaning, and reducing the influence of these dust on subsequent detection. When the detection ring 14 is lifted, the air bag strips 17 are in contact with the tooth tip surface of the gear 3, and the air bag strips 17 are always in an expanded state. At the same time, piezoelectric sensors are arranged on the surfaces of the air bag strips 17, and the piezoelectric sensors are electrically connected with the controller. At the same time, as shown in Figure 6 and Figure 8 In this embodiment, due to the existence of the detection ring 14 and the lifting table 13, the installation position and model of the laser emitter-receiver switch 8 are adjusted accordingly, and the model of the laser emitter-receiver switch 8 can be preferably M3-T2000N, so that the laser emitter-receiver switch 8 can be embedded in the top end of the lifting table 13 and correspond to the through grooves 15, so that the light of the laser emitter-receiver switch 8 can be blocked by the tooth tip of the gear 3, thereby not affecting the alignment of the gear 3 and the reflector 5.

[0060] The specific movement is as follows:

[0061] In the initial state, the gear 3 to be detected is fixed by the spline of the cooperating column 2, and the tooth tip of the gear 3 corresponds to the through groove 15 of the detection ring 14 one by one (the number of through grooves 15 matches the number of teeth of the gear 3). At this time, the detection ring 14 is located below the gear 3, the air bag strips 17 are in an expanded state but not in contact with the tooth tip, and the distance between the scraping blocks 16 and the tooth tip is about 2-3 mm.

[0062] When the detection ring 14 rises, the controller drives the lifting platform 13 (annular structure) to move upward, driving the detection ring 14 to rise synchronously. First, the scraping block 16 (the surface is covered with a sponge layer) inside the through slot 15 contacts the tooth top surface of the gear 3 before the air bag strip 17. As the detection ring 14 continues to rise, the scraping block 16 slides along the tooth top surface. By using the flexible adsorption effect of the sponge layer, dust, iron filings and other impurities on the tooth top surface are removed, avoiding the fact that stains cause the air bag strip 17 to not contact the tooth top, and ensuring that the pressure signal is accurate. As the detection ring 14 continues to rise, the air bag strip 17 (always inflated) inside the through slot 15 immediately comes into close contact with the tooth top surface. Since the air bag strip 17 is elastic, it will adaptively deform along the shape of the tooth top surface during the rising process (if the tooth top is flat, the air bag strip 17 uniformly contacts the tooth top surface, the pressure distribution is stable, and the detection value is the standard value A).

[0063] However, if the tooth top has protrusions (such as burrs, machining allowances, generally small in volume), the contact area between these protrusions and the surface of the air bag strip 17 is small, forming a local high pressure on the surface of the air bag strip 17; if the tooth top has depressions (such as material defects, wear), the corresponding position of the air bag strip 17 is subjected to reduced force (relative to the region not in contact with the tooth top surface).

[0064] Then, the piezoelectric sensor on the surface of the air bag strip 17 converts the pressure change into an electrical signal (piezoelectric effect: the greater the pressure, the higher the output voltage) in real time. The controller receives the continuous voltage signal and converts it into a pressure curve. By analyzing the fluctuation amplitude of the curve (such as the peak corresponding to the protrusion, the valley corresponding to the depression), the distribution characteristics of the tooth top flatness can be quantified. In addition, for the judgment of the above two cases (protrusion and depression), by comparing with the standard value A, it is obtained that greater than the standard value A is a protrusion, and less than the standard value A is a depression (for example: when the tooth top has a 0.1mm high burr, the output voltage of the piezoelectric sensor at the corresponding position will be increased by 30%-50% compared with the flat area; when the tooth top has a 0.05mm deep pit, the voltage signal at the corresponding position will be reduced by 20%-30%).

[0065] Secondly, the controller compares the pressure signal of the piezoelectric sensor with the tooth top image collected by the image sensor 7 in Example 1 (for example: the "step-shaped protrusion" shown in the image should coincide with the peak position of the pressure curve), and through double data cross verification, the reliability of the detection result is improved. After the detection is completed, the lifting platform 13 drives the detection ring 14 to descend to the initial position, waiting for the next detection.

[0066] Obviously, the above examples are only examples for the purpose of clear illustration, and are not a limitation on the embodiments. For those skilled in the art, other different forms of changes or variations can be made on the basis of the above description. Here, it is not necessary and impossible to exhaust all the embodiments. The obvious changes or variations derived therefrom are still within the protection scope of the present invention.

Claims

1. A device for detecting the addendum flatness of a gear, characterized by, Including holding platform (1) and controller, holding platform (1) is equipped with cooperation column (2), cooperation column (2) is used to install the gear (3) to be detected; Holding platform (1) is equipped with reflection frame (4), reflection frame (4) surrounds cooperation column (2), the inside of reflection frame (4) is equipped with several inclined mirrors (5); The top of reflection frame (4) is equipped with crosspiece (6), the bottom of crosspiece (6) is fixedly connected with image sensor (7), image sensor (7) is used to gather the image of each mirror (5) reflection; The controller is used to obtain the image collected by image sensor (7), and the controller is used to judge whether the gear (3) tooth top is flat based on image recognition; The top of holding platform (1) is fixedly connected with lifting platform (13) electrically connected with the controller, the output end of lifting platform (13) is fixedly connected with detection ring (14), the inner ring of detection ring (14) is opened with several through slots (15) corresponding with gear (3) teeth, the inner side of through slot (15) is all bonded with several air bag strips (17) from top to bottom and the air bag strip (17) is always in the inflation state;The inner side of through slot (15) is all fixedly connected with scraping block (16), and the scraping block (16) is all located above the air bag strip (17), the surface of scraping block (16) is all bonded with sponge layer, and the scraping block (16) is all close to the top of detection ring (14) and is inclinedly arranged;The surface of air bag strip (17) is all provided with piezoelectric sensor, and the piezoelectric sensor is all electrically connected with the controller.

2. The addendum flatness detection apparatus for a gear according to claim 1, characterized by, The inclination angle of mirror (5) is 45 degrees, and the mirror (5) is detachably connected with the inside of reflection frame (4).

3. The addendum flatness detection apparatus for a gear according to claim 2, characterized by, The inside of reflection frame (4) is provided with a magic tape surface, the back of mirror (5) is provided with a magic tape hook surface, and mirror (5) and reflection frame (4) are detachably connected by magic tape surface and magic tape hook surface.

4. The addendum flatness detection apparatus for a gear according to claim 3, characterized by The imaging range width of gear (3) tooth top of mirror (5) corresponds to the width of gear (3) tooth top.

5. The addendum flatness detection apparatus for a gear according to claim 4, characterized by The holding platform (1) and the crosspiece (6) are provided with a pair of laser emitting and receiving switches (8), and the monitoring range of the laser emitting and receiving switch (8) is aligned with the edge of the imaging range of the mirror (5).

6. The addendum flatness detection apparatus for a gear according to claim 5, characterized by The cooperation column (2) is provided with a motor for driving the cooperation column (2) to rotate, and the controller is used to control the motor to drive the cooperation column (2) to rotate, and the laser emitting and receiving switch (8) is used to judge whether the gear (3) tooth reaches the monitoring range of the laser emitting and receiving switch (8), and the controller controls the motor to brake if it reaches.

7. The addendum flatness detection apparatus for a gear according to claim 6, characterized by The bottom of crosspiece (6) is provided with telescopic rod (9), and the end of telescopic rod (9) away from crosspiece (6) is fixedly connected with blowing cylinder (10).

8. The addendum detection apparatus for a gear according to claim 7, characterized by The bottom of crosspiece (6) is provided with annular slide rail (11), and telescopic rod (9) is slidingly connected on the slide rail, and the slide rail is provided with a driving member for driving telescopic rod (9) to slide along the slide rail.

9. The addendum flatness detection apparatus for a gear according to claim 8, characterized by, The end of telescopic rod (9) away from crosspiece (6) is fixedly connected with wiping block (12), and wiping block (12) is located at the back of blowing cylinder (10), and wiping block (12) is used to clean mirror (5) when driving member drives telescopic rod (9) to slide.

10. The addendum detection apparatus for a gear according to claim 9, wherein The controller is internally provided with a trained convolutional neural network, the convolutional neural network is trained based on a plurality of images of steps, recesses, collapses, scratches and patterns of tooth crests of gears (3), and the convolutional neural network is used for inputting images collected by an image sensor (7) and framing and marking steps, recesses, collapses, scratches and patterns in the images.

Citation Information

Patent Citations

  • A gear tooth top flatness detection device

    CN119334230B

  • Bottom imaging device based on mirror reflection

    CN214953075U

  • Device for detecting toughness of part design sample material

    CN222124999U