AI visual inspection equipment for flywheel defect detection

Abstract

The invention relates to the technical field of flywheel detection, and discloses an AI visual detection device for flywheel defect detection, comprising a station for bearing a flywheel, the station being capable of reciprocating along a first preset path, a first driving assembly for driving the station to reciprocate along the first preset path, and a detection mechanism for detecting different defect information of the flywheel, according to the flywheel defect detection device, the station reciprocates on the first preset path and passes through the multiple detection mechanisms, detection of different kinds of defect information of the flywheel is carried out at the different detection mechanisms, detection of multiple types of defects and specific parameters of the flywheel can be achieved, multi-mechanism cooperative calibration is achieved, the detection precision is improved, and in addition, the detection efficiency is improved. The whole detection process is completed on the same moving path, the flywheel does not need to be manually moved, the turnover time consumption is reduced, continuous circulation of detection, discharging and feeding can be achieved through reciprocating movement of the stations, and the detection efficiency is improved.

Description

Technical Field

[0001] The present invention relates to the technical field of flywheel detection, and in particular to an AI vision detection device for flywheel defect detection. Background Art

[0002] A flywheel is a mechanical device for storing and releasing energy, usually a disc-shaped part with a large moment of inertia. The quality of the flywheel directly affects its use effect. After the flywheel is processed, it needs to undergo quality inspection, and only when the quality is qualified can it leave the factory.

[0003] The flywheel includes structures such as a rim, a hub, pins, bearings, and webs. There are multiple teeth on the rim. Therefore, when detecting, various defects of each structure of the flywheel need to be detected, such as defects on the upper surface, lower surface, and side surface, the size of the pins, the number of teeth, the height, etc. The existing flywheel defect detection devices can only detect single-type defects or specific parameters of the flywheel, or detect through multiple independent detection devices, which requires moving the flywheel multiple times and it is difficult to complete comprehensive detection in an integrated device, resulting in inaccurate or imperfect detection results. Summary of the Invention

[0004] The purpose of the present invention is to provide an AI vision detection device for flywheel defect detection to solve the problems in the prior art.

[0005] The present invention provides an AI vision detection device for flywheel defect detection, including:

[0006] A work station for carrying the flywheel, and the work station can reciprocate along a first preset path;

[0007] A first driving component for driving the work station to reciprocate along the first preset path;

[0008] A detection mechanism, including a first detection component, a second detection component, and a third detection component arranged in sequence on the first preset path, and the detection mechanism is used to detect different defect information of the flywheel;

[0009] A workbench for carrying the work station, the first driving component, and the detection mechanism.

[0010] An AI vision detection device for flywheel defect detection as described above, wherein, preferably, the work station includes a carrier and a plurality of limiting components, and the plurality of limiting components are annularly spaced on the carrier for limiting the flywheel on the carrier;

[0011] The limiting component includes a sliding member and a baffle. A chute is formed on the supporting bracket. The sliding member can move within the chute. The baffle protrudes from the sliding member. The flywheel is placed on the sliding member. The sliding member is moved to make the baffle abut against the side wall of the flywheel.

[0012] An AI vision detection device for flywheel defect detection as described above, wherein, preferably, the first driving component includes a first guiding member and a first driving member. A first guiding and cooperating member is provided at the bottom of the working station. The first guiding and cooperating member is in guiding cooperation with the first guiding member. The first driving member drives the first guiding and cooperating member to move on the first guiding member.

[0013] An AI vision detection device for flywheel defect detection as described above, wherein, preferably, the first detection component includes a second driving component and a displacement sensor. A first support frame is provided on the workbench. The second driving component is arranged on the side of the first support frame away from the first driving component. The second driving component drives the displacement sensor to move along a second preset path. The displacement sensor performs multi-point detection on the first defect of the flywheel during the movement along the second preset path.

[0014] An AI vision detection device for flywheel defect detection as described above, wherein, preferably, the second driving component includes a second guiding member and a second driving member. A second guiding and cooperating member is provided on the displacement sensor. The second guiding and cooperating member is in guiding cooperation with the second guiding member. The second driving member drives the second guiding and cooperating member to move on the second guiding member.

[0015] An AI vision detection device for flywheel defect detection as described above, wherein, preferably, the AI vision detection device further includes a third driving component arranged on the first preset path. When the working station moves to the position corresponding to the first detection component, the third driving component drives the flywheel to rise or fall and rotate at a preset angle after rising.

[0016] An AI vision detection device for flywheel defect detection as described above, wherein, preferably, the third driving component includes a telescopic driving member and a rotary driving member. A first channel is formed on the working station. The telescopic driving member is arranged on the side of the workbench away from the first detection component. The rotary driving member is arranged at the output end of the telescopic driving member. The output end of the rotary driving member can pass through the first channel and abut against the bottom of the flywheel. The telescopic driving member is used to drive the rotary driving member to lift and lower. The rotary driving member is used to drive the flywheel to rotate.

[0017] An AI vision detection device for flywheel defect detection as described above, wherein, preferably, the second detection component includes a first AI camera and a fourth driving component. The lens of the first AI camera faces the upper surface of the flywheel and is used to detect the defect information on the upper surface of the flywheel. The fourth driving component drives the first AI camera to approach or move away from the flywheel.

[0018] An AI vision detection device for flywheel defect detection as described above, wherein, preferably, the third detection component includes a second AI camera. The second AI camera is arranged below the workbench, and the lens of the second AI camera faces the lower surface of the flywheel and is used to detect the defect information on the bottom surface of the flywheel.

[0019] An AI vision detection device for flywheel defect detection as described above, wherein, preferably, the AI vision detection device further includes a fourth detection component. The fourth detection component includes a third AI camera arranged on the workbench. The lens of the third AI camera faces the side surface of the flywheel and is used to detect the defect information on the side surface of the flywheel.

[0020] Compared with the prior art, in the present invention, the workbench reciprocates on the first preset path and passes through multiple detection mechanisms respectively, and detects different defect information of the flywheel at different detection mechanisms, so as to realize the detection of multiple types of defects and specific parameters of the flywheel. In addition, multiple mechanisms cooperate for calibration to improve the detection accuracy. Moreover, the entire detection process is completed on the same moving path, without manual movement of the flywheel, reducing the turnover time. The reciprocating movement of the workbench can realize the continuous cycle of "detection - blanking - loading", improving the detection efficiency. Description of the Drawings

[0021] Figure 1 is a perspective view of the AI vision detection device for flywheel defect detection provided by the embodiment of the present invention;

[0022] Figure 2 is a top view of the AI vision detection device for flywheel defect detection provided by the embodiment of the present invention;

[0023] Figure 3 is a side view of the AI vision detection device for flywheel defect detection provided by the embodiment of the present invention;

[0024] Figure 4 is a perspective view of the first driving component and the workbench provided by the embodiment of the present invention;

[0025] Figure 5 is a perspective view of the workbench provided by the embodiment of the present invention;

[0026] Figure 6It is a perspective view of the first detection component provided by an embodiment of the present invention;

[0027] Figure 7 It is a perspective view of the third driving component provided by an embodiment of the present invention;

[0028] Figure 8 It is a perspective view of the second detection component provided by an embodiment of the present invention.

[0029] Explanation of reference numerals:

[0030] 10 - work station, 11 - carrier, 110 - chute, 12 - limiting component, 120 - sliding part, 121 - baffle, 13 - first guiding and fitting part, 14 - first channel;

[0031] 20 - first driving component, 21 - first guiding part, 22 - first driving part;

[0032] 30 - first detection component, 31 - second driving component, 310 - second guiding part, 311 - second driving part, 32 - displacement sensor, 320 - second guiding and fitting part

[0033] 40 - second detection component, 41 - first AI camera, 42 - fourth driving component,

[0034] 50 - third detection component;

[0035] 60 - fourth detection component;

[0036] 70 - third driving component, 71 - telescopic driving part, 710 - telescopic rod, 711 - first air cylinder, 712 - first mounting bracket, 72 - rotary driving part, 720 - rotary platform, 721 - servo motor, 722 - magnetic attraction structure, 723 - second mounting bracket.

[0037] 80 - workbench, 81 - first support frame, 82 - loading rack, 83 - central control component, 84 - display panel;

[0038] 90 - flywheel. Detailed implementation manners

[0039] The embodiments described below with reference to the accompanying drawings are exemplary and are only used to explain the present invention, and should not be construed as a limitation to the present invention.

[0040] See Figures 1 - 3 As shown, the present invention provides an AI vision detection device for flywheel defect detection, including a work station 10, a first driving component 20, a detection mechanism, and a workbench 80, wherein:

[0041] Station 10 is used to carry the flywheel 90. Station 10 can reciprocate along a first preset path. The first driving component 20 is used to drive station 10 to reciprocate along the first preset path. The detection mechanism includes a first detection component 30, a second detection component 40, and a third detection component 50 sequentially arranged on the first preset path. The detection mechanism is used to detect different defect information of the flywheel 90. The workbench 80 is used to carry station 10, the first driving component 20, the detection mechanism, etc. There are multiple detection points on the first preset path of this application. A detection mechanism is correspondingly arranged at each detection point. When the flywheel 90 arrives at the corresponding detection point and stays, after the detection mechanism collects information, it moves to the next detection point, which can comprehensively cover various possible defects of the flywheel 90, improve the detection efficiency, and there is no need to manually take and place the flywheel 90 on different test devices for testing.

[0042] See Figures 1 - 3 As shown, in this embodiment, station 10 is set to one. After the detection is completed, it is reset to the starting end of the first preset path. The operator only needs to load and unload at the loading position of the equipment, which reduces the workload. A discharge rack 82 is arranged on the side of the workbench 80. A plurality of partitions are arranged at intervals on the discharge rack 82 and can be used to place defective products.

[0043] See Figures 1 - 5 As shown, in order to adapt to flywheels 90 with different diameters, station 10 includes a carrier 11 and a plurality of limiting components 12. The plurality of limiting components 12 are annularly arranged at intervals on the carrier 11 and are used to limit the flywheel 90 on the carrier 11. The limiting component 12 includes a sliding member 120 and a baffle 121. A chute 110 is formed on the carrier 11. The sliding member 120 can move in the chute 110. The baffle 121 protrudes from the sliding member 120. The flywheel 90 is placed on the sliding member 120. The sliding member 120 is moved so that the baffle 121 abuts against the side wall of the flywheel 90. In the embodiment provided in this application, the limiting component 12 is set to four. The sliding member 120 can be a slider. The slider is slidably matched with the chute 110, and an adjusting bolt is provided on the slider. By adjusting the four sliders, the internal space formed by the enclosure of the four baffles 121 can be changed. When changing the model of the flywheel 90, the center of the flywheel 90 is aligned with the center of station 10. This method can be realized by the prior art and will not be elaborated here. The flywheel 90 is placed on the slider. The sliders are moved so that the baffles 121 all press against the outer edge of the flywheel 90. When adjusted to the appropriate position, the slider and the chute 110 are fixed with an adjusting nut. After that, the flywheels 90 of this model can automatically align the center, improving the placement efficiency and the accuracy of the detection effect.

[0044] See Figure 4As shown, in order to move the work station 10 along the first preset path, the first driving assembly 20 includes a first guiding member 21 and a first driving member 22. A first guiding and mating member 13 is provided at the bottom of the work station 10. The first guiding and mating member 13 is in guiding cooperation with the first guiding member 21, and the first driving member 22 drives the first guiding and mating member 13 to move on the first guiding member 21. In this embodiment, the first guiding member 21 is a linear guide rail, the first guiding and mating member 13 is a slider, and the first driving member 22 is a motor. In this application, two parallel linear guide rails are provided, and sliders are provided at the four corners at the bottom of the work station 10 to increase the smoothness of the movement of the work station 10. As another implementation manner of this application, the first guiding member 21 can also be a lead screw or the like.

[0045] Referring to Figure 6 As shown, the first detection assembly 30 includes a second driving assembly 31 and a displacement sensor 32. A first support frame 81 is provided on the workbench 80. The second driving assembly 31 is provided on the side of the first support frame 81 away from the first driving assembly 20. The second driving assembly 31 drives the displacement sensor 32 to move along the second preset path. The displacement sensor 32 performs multi-point detection on the first defect of the flywheel 90 during the movement along the second preset path. When the flywheel 90 reaches the corresponding position of the first detection assembly 30, it stops. At this time, the displacement sensor 32 moves along the second preset path. The purpose is to measure the pin, bearing installation size and gear ring height of the flywheel 90 respectively on the movement path, so as to collect more comprehensive defect information. In this embodiment, the model of the displacement sensor 32 can be the TR-0050-0.5 displacement sensor of NOVOTECHNIK Company or other models, which is not limited here.

[0046] In this embodiment, referring to Figure 6 As shown, the second driving assembly 31 includes a second guiding member 310 and a second driving member 311. A second guiding and mating member 320 is provided on the displacement sensor 32. The second guiding and mating member 320 is in guiding cooperation with the second guiding member 310, and the second driving member 311 drives the second guiding and mating member 320 to move on the second guiding member 310. In this embodiment, the second guiding member 310 is a linear guide rail, the second guiding and mating member 320 is a slider, and the second driving member 311 is a motor.

[0047] Referring to Figure 1 and Figure 3As shown, since the flywheel 90 is circular, in order to collect more comprehensive defect information, it is best for the displacement sensor 32 to collect multiple groups of data at intervals along the circumference of the flywheel 90. However, if the second preset path is circular, it will block the upper surface of the flywheel 90, which is not conducive to subsequent collection of surface defects, and the circular guide will increase costs. Therefore, the second preset path is preferably set to a straight line. In order to collect multiple groups of data along the circumference of the flywheel 90, in this embodiment, the AI vision detection device further includes a third driving component 70, which is arranged on the first preset path. When the station 10 moves to the position corresponding to the first detection component 30, the third driving component 70 drives the flywheel 90 to rise or fall, and rotates at a preset angle after rising. By driving the flywheel 90 to rise or fall through the third driving component 70, the dimensions or thicknesses of more structures of the flywheel 90 can be collected. When collecting the thickness, for example, the thickness of the gear ring of the flywheel 90, another displacement sensor 32 corresponding to the output end of the displacement sensor 32 can also be arranged below the flywheel 90. The two displacement sensors 32 contact the flywheel 90 at the same time, and the thicknesses at multiple positions can be measured. Of course, in order to facilitate the displacement sensor 32 below to still be able to contact the surface of the flywheel 90 after the flywheel 90 rises, a corresponding driving structure can be configured. After the flywheel 90 rises, it rotates at a preset angle. During the rotation of the flywheel 90 along the axis, the displacement sensor 32 can collect multiple groups of data at multiple positions. The preset angle can be 180 degrees or 90 degrees.

[0048] See Figure 7As shown, in this embodiment, the third driving assembly 70 includes a telescopic driving member 71 and a rotary driving member 72. A first channel 14 is formed on the work station 10. The telescopic driving member 71 is arranged on the side of the workbench 80 away from the first detection assembly 30. The rotary driving member 72 is arranged at the output end of the telescopic driving member 71. The output end of the rotary driving member 72 can pass through the first channel 14 and abut against the bottom of the flywheel 90. The telescopic driving member 71 is used to drive the rotary driving member 72 to lift and lower, and the rotary driving member 72 is used to drive the flywheel 90 to rotate. The diameter of the first channel 14 is larger than the diameter of the flywheel 90. At least a part of the sliding member 120 extends into the projected area of the first channel 14, so that while supporting the flywheel 90, the lower surface of the flywheel 90 can be blocked as little as possible. In this embodiment, the telescopic driving member 71 includes a telescopic rod 710 and a first cylinder 711. The telescopic driving member 71 is connected to the workbench 80 through a first mounting bracket 712. The rotary driving member 72 includes a rotary platform 720 and a servo motor 721. The rotary platform 72 is connected to the output end of the telescopic rod 710 through a second mounting bracket 723. The driving modes of the telescopic driving member 71 and the rotary driving member 72 can be realized by existing technologies and are not limited herein. It should be noted that in this embodiment, the diameter of the rotary platform is smaller than the diameter of the flywheel 90, and the purpose is to pass through the first channel 14 and abut against the flywheel 90. In order to increase the stability of the flywheel 90 during lifting and rotating, a magnetic attraction structure 722 can be arranged at the output end of the rotary driving member 72. The magnetic attraction structure 722 can be a strong magnet such as a magnet, and the purpose is to stably attract the flywheel 90.

[0049] In this embodiment, when the work station 10 moves to the first detection assembly 30, the first cylinder 711 drives the telescopic rod 710 to lift the second mounting bracket 723 and the rotary driving member 72 thereon until the rotary platform 720 abuts against the bottom of the flywheel 90. The magnetic attraction structure 722 adsorbs it, and drives the telescopic rod 710 to continue to lift the flywheel 90 by a certain height to make the flywheel 90 suspended. The servo motor 721 drives the rotary platform 720 to rotate at a preset angle, and the first detection assembly 30 detects the flywheel 90. After the detection is completed, the first cylinder 711 drives the telescopic rod 710 to lower the second mounting bracket 723 and the rotary driving member 72 thereon. When the flywheel 90 abuts against the work station 10, the magnetic attraction structure 722 cancels the adsorption of the flywheel 90, and the telescopic driving member 71 only needs to lift the flywheel 90 to be suspended from the work station 10.

[0050] In another embodiment, the output end of the third driving assembly 70 can also grab or adsorb the flywheel 90 from above it. It also includes a telescopic driving member 71 and a rotary driving member 72. A gripper or a suction cup etc. is arranged at the output end of the rotary driving member 72, and its specific structure is not limited herein.

[0051] See Figure 8As shown in the figure, the second detection component 40 includes a first AI camera 41 and a fourth driving component 42. The lens of the first AI camera 41 faces the upper surface of the flywheel 90 and is used to detect the defect information on the upper surface of the flywheel 90. The fourth driving component 42 drives the first AI camera 41 to approach or move away from the flywheel 90. In this embodiment, visual detection is used to collect the surface defects of the flywheel 90. Specifically, an AI camera is used, and the model can be MV-CH250-90GM, etc. The functions such as intelligent recognition and image processing of the AI camera can be used to detect the number of teeth, pins, bearings, and surface defects.

[0052] See Figure 8 As shown in the figure, the fourth driving component 42 in this embodiment adopts a structure such as a guide rail or a lead screw, which can be the same as the second driving component 31, and its specific structure is not limited here. Due to the different sizes of the flywheel 90, if the position of the first AI camera 41 is fixed, it may not be able to take a comprehensive picture or the picture may be blurred when obtaining the surface information. Therefore, it is designed that the first AI camera 41 can move up and down, and for flywheels 90 of different sizes, they can all be photographed at an optimal height. The selection of the optimal shooting height can be set by the system or an induction device can be added, etc., which is not limited here. It should be noted that the flywheel 90 can be photographed while being detected by the first detection component 30. At this time, the flywheel 90 is rotating, and the first AI camera 41 can still take pictures of multiple positions and analyze them through integration, thereby reducing the detection time.

[0053] In this embodiment, see Figures 1 - 3 As shown in the figure, the third detection component 50 includes a second AI camera. The second AI camera is arranged below the workbench 80, and the lens of the second AI camera faces the lower surface of the flywheel 90 and is used to detect the defect information on the bottom surface of the flywheel 90. The second AI camera can be arranged at the end of the first preset path. When the flywheel 90 descends to the station 10 after being detected by the first detection component 30 and the second detection component 40, it is moved to the end of the first preset path by the first driving component 20.

[0054] Furthermore, see Figures 1 - 3 As shown in the figure, the AI vision detection device further includes a fourth detection component 60. The fourth detection component 60 includes a third AI camera, which is arranged on the workbench 80, and the lens of the third AI camera faces the side surface of the flywheel 90 and is used to detect the defect information on the side surface of the flywheel 90. The third AI camera can take pictures of the side information of multiple areas during the rotation of the flywheel 90, so as to detect information such as the engraved lines and the number of side holes of the flywheel 90.

[0055] See Figure 1As shown, the AI vision detection device further includes a central control component 83, a display panel 84, etc., which are used to control the device and display information such as detection data and results. It should be noted that the circuits and controls involved in the present invention are all prior arts and will not be elaborated here.

[0056] Based on the above embodiments, the working principle of an AI vision detection device for flywheel defect detection in the present application is as follows:

[0057] The flywheel 90 is manually placed on the work station 10. By adjusting the limit component 12 to adapt to flywheels 90 of different sizes, the device is started. The work station 10 moves along the first guiding member 21 to the position of the first detection component 30. The lifting driving member drives the rotating driving member 72 to rise. The output end of the rotating driving member 72 passes through the first channel 14 and abuts against the bottom of the flywheel 90, lifting the flywheel 90 to a certain height. The rotating driving member 72 drives the flywheel 90 to rotate and pause at a preset angle. During the pause, the displacement sensor 32 moves along the second preset path and detects relevant defects such as pins, gear rings, bearings, etc. Specifically, it can include the plane and height information of the pin position, the judgment of the gear ring position, etc. At the same time, the first AI camera 41 moves to the optimal shooting height under the drive of the fourth driving component 42 to shoot various possible defects on the upper surface of the flywheel 90. The third AI camera shoots the defect information of multiple sides of the flywheel 90 during its rotation. After completion, the flywheel 90 descends to the work station 10 and continues to move to the second AI camera to shoot the defects on its lower surface. After all detections are completed, the work station 10 drives the flywheel 90 to move to the starting end of the first preset path, and the operator classifies the qualified products and defective products according to the detection results and conducts the detection of the next flywheel 90.

[0058] The structure, features and function effects of the present invention have been described in detail based on the embodiments shown in the drawings. The above are only the preferred embodiments of the present invention, but the present invention is not limited to the scope defined by the drawings. Any changes made according to the concept of the present invention, or modified into equivalent embodiments with equivalent changes, still within the spirit covered by the description and the drawings, shall be within the protection scope of the present invention.

Claims

1. An AI vision inspection device for flywheel defect detection, characterized in that, Comprising: A work station for carrying a flywheel, the work station being reciprocally movable along a first preset path; A first driving assembly for driving the work station to reciprocally move along the first preset path; A detection mechanism including a first detection component, a second detection component, and a third detection component sequentially arranged on the first preset path, the detection mechanism being used to detect different defect information of the flywheel; A workbench for carrying the work station, the first driving assembly, and the detection mechanism.

2. The AI vision detection device for flywheel defect detection according to claim 1, characterized in that The work station includes a carrier bracket and a plurality of limiting components, the plurality of limiting components being annularly spaced on the carrier bracket for limiting the flywheel on the carrier bracket; The limiting component includes a sliding member and a baffle, a sliding groove is formed on the carrier bracket, the sliding member is movable in the sliding groove, the baffle protrudes from the sliding member, the flywheel is placed on the sliding member, and the sliding member is moved to make the baffle abut against the side wall of the flywheel.

3. The AI vision detection device for flywheel defect detection according to claim 1, characterized in that, The first driving assembly includes a first guiding member and a first driving member, a first guiding and mating member is provided at the bottom of the work station, the first guiding and mating member is in guiding and mating with the first guiding member, and the first driving member drives the first guiding and mating member to move on the first guiding member.

4. The AI vision detection device for flywheel defect detection according to claim 1, wherein, The first detection component includes a second driving assembly and a displacement sensor, a first support frame is provided on the workbench, the second driving assembly is provided on the side of the first support frame away from the first driving assembly, the second driving assembly drives the displacement sensor to move along a second preset path, and the displacement sensor performs multi-point detection on the first defect of the flywheel during the movement along the second preset path.

5. The AI vision detection device for flywheel defect detection according to claim 4, characterized in that, The second driving assembly includes a second guiding member and a second driving member, a second guiding and mating member is provided on the displacement sensor, the second guiding and mating member is in guiding and mating with the second guiding member, and the second driving member drives the second guiding and mating member to move on the second guiding member.

6. The AI vision detection device for flywheel defect detection according to claim 1, characterized in that, The AI vision detection device further includes a third driving assembly arranged on the first preset path. When the work station moves to the position corresponding to the first detection component, the third driving assembly drives the flywheel to rise or fall and rotate at a preset angle after rising.

7. The AI vision detection device for flywheel defect detection according to claim 6, characterized in that, The third driving assembly includes a telescopic driving member and a rotating driving member. A first channel is formed on the work station. The telescopic driving member is provided on the side of the workbench away from the first detection component, the rotating driving member is provided at the output end of the telescopic driving member, the output end of the rotating driving member can pass through the first channel and abut against the bottom of the flywheel, the telescopic driving member is used to drive the rotating driving member to rise and fall, and the rotating driving member is used to drive the flywheel to rotate.

8. The AI vision detection device for flywheel defect detection according to claim 1, characterized in that, The second detection component includes a first AI camera and a fourth driving assembly. The lens of the first AI camera faces the upper surface of the flywheel for detecting defect information on the upper surface of the flywheel, and the fourth driving assembly drives the first AI camera to approach or move away from the flywheel.

9. The AI vision detection device for flywheel defect detection according to claim 1, wherein, The third detection component includes a second AI camera. The second AI camera is provided below the workbench, and the lens of the second AI camera faces the lower surface of the flywheel for detecting defect information on the bottom surface of the flywheel.

10. The AI vision detection device for flywheel defect detection according to claim 1, characterized in that, The AI vision detection device further includes a fourth detection component, and the fourth detection component includes a third AI camera which is disposed on the workbench. The lens of the third AI camera faces the side surface of the flywheel and is used for detecting the defect information on the side surface of the flywheel.

Images