Intelligent connecting piece identifying and screening system suitable for automatic assembly

Abstract

The invention relates to the technical field of connector screening, in particular to an intelligent connector identifying and screening system suitable for automatic assembly.The system comprises an external module, a feeding module, a conveying module, a detecting module and a screening module.The detecting module comprises a detecting unit, and the conveying module comprises a conveying base and a conveying unit; the conveying unit comprises a conveying rotary disc and conveying wire ropes, the conveying rotary disc is rotationally arranged on the conveying base, the conveying wire ropes are movably arranged on the periphery of the conveying rotary disc, and the connecting piece is arranged between the two conveying wire ropes. The conveying turntable drives the paired wire ropes to form an adjustable clamping channel, the posture of the connecting piece is automatically corrected in the conveying process, and meanwhile, a detection coordinate system is established by the boundaries of the wire ropes, so that the surface characteristics of the connecting piece and a datum line form a clear comparison relationship; through the synergistic effect of mechanical conveying and optical detection, the problem of misjudgment caused by workpiece deviation in traditional static detection is effectively solved.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of connector screening, and particularly relates to a connector intelligent identification screening system suitable for automatic assembly. BACKGROUND

[0002] At present, in the aviation manufacturing industry, automatic assembly is gradually replacing the traditional manual assembly mode. Among them, the automatic drilling and riveting system as the key equipment of automatic assembly undertakes a large number of installation tasks of connectors. The connectors for automatic assembly enter the work station by automatic feeding mode. When the connectors are filled, due to the large number of connectors, it is difficult to carefully screen each connector. The connectors that are not screened have defects or inconsistent models. After assembly, there are problems such as unqualified or unattractive appearance quality, which not only affects the overall quality and safety of the aviation parts, but also may cause production delay and cost increase.

[0003] At present, in the automatic drilling and riveting process, manual screening is a common connector screening method, but this method has many disadvantages. First, when manually screening, the rivet defects are often very subtle and difficult to detect with the naked eye, so the phenomenon of missed screening frequently occurs, which makes a large number of unqualified connectors flow into the subsequent production link, seriously affecting the product quality. Secondly, the aviation manufacturing industry has a large production scale, and the number of connectors in a single batch is large. Relying entirely on manual screening, the speed is slow, which is difficult to match the production rhythm of the automatic drilling and riveting system, resulting in a decrease in overall production efficiency. Finally, for the same brand but different identification connectors, the existing technology has not established a perfect differentiation system, and can only be mixed, which cannot guarantee the consistency and standardization of the products, increasing the quality risk in the production process. In addition, the existing technology is difficult to accurately identify the small scratches on the rivet head surface and the missing corners.

[0004] Therefore, it is urgent to design a full-automatic, efficient and accurate connector screening system for the automatic drilling and riveting system. The system is composed of a vibration feeding mechanism, a visual detection mechanism, a conveying mechanism, a material separation mechanism and a system outer frame. The conveying mechanism adopts a design method combining a flexible conveying belt and a width adjustment mechanism to realize the sharing of the same conveying belt by different types of connectors. The visual detection mechanism for screening out defective connectors is embedded with deep learning function to reduce product defects caused by defective connectors and improve product quality. SUMMARY

[0005] The main purpose of the present application is to provide a connector intelligent identification screening system suitable for automatic assembly, which aims to solve the problem that the screening system in the prior art cannot identify the small scratches on the rivet head surface and the missing corners.

[0006] In order to achieve the above object, the application provides a connecting piece intelligent identification and screening system suitable for automatic assembly, which is used for detecting and screening connecting pieces, and comprises An external module, which comprises a system outer frame, and a control unit is arranged in the system outer frame; A feeding module, which is arranged in the system outer frame and is used for completing the arrangement and conveying of connecting pieces; A conveying module, which is arranged in the system outer frame and is used for completing the position adjustment of connecting pieces; A detection module, which is arranged in the system outer frame and is used for completing the identification and detection of connecting pieces in the process of position adjustment of connecting pieces; A screening module, which is arranged in the system outer frame and is used for completing the screening of connecting pieces according to the results of identification and detection; The detection module comprises a detection unit, the conveying module comprises a conveying seat and a pair of conveying units, the conveying units comprise a conveying turntable and a conveying wire rope, the conveying turntable is rotationally arranged on the conveying seat, the conveying wire rope is movably arranged on the outer periphery of the conveying turntable, the connecting pieces are arranged between the two conveying wire ropes, and the position adjustment of the connecting pieces is completed through the rotation of the conveying turntable, and the detection unit is used for completing the identification and detection of the connecting pieces with the boundary information of the conveying wire rope as the reference.

[0007] Further, the external module further comprises a display unit and an interaction unit, the display unit is used for displaying screening information, the screening information comprises the abnormal number and the abnormal type of the connecting pieces, and the interaction unit is used for completing information interaction.

[0008] Further, a placing rack is arranged in the system outer frame, the feeding module comprises a vibration feeding unit and a linear feeding unit, the vibration feeding unit and the linear feeding unit are both arranged on the placing rack, and the horizontal height of the input end of the linear feeding unit is higher than that of the output end.

[0009] Further, the detection module comprises a defect detection unit, the defect detection unit is arranged above the conveying wire rope, and the defect detection unit comprises a defect detection support, a defect detection camera, a defect detection lens and a defect detection light source, the defect detection support is arranged in the system outer frame, the defect detection camera and the defect detection lens are arranged on the defect detection support in cooperation, and the defect detection light source is arranged on the defect detection support and placed in front of the defect detection lens.

[0010] Further, the detection module further comprises a size detection unit arranged at the side of the conveying wire rope, the size detection unit comprises a size detection support, a size detection camera, a size detection lens and a size detection light source, the size detection support is arranged in the system outer frame, the size detection camera is arranged on the size detection support after being matched with the size detection lens, and the size detection light source is arranged in front of the size detection lens, so that the conveying wire rope is located between the size detection lens and the size detection light source.

[0011] Further, the detection module further comprises a sensing support and a sensing unit, the sensing support is arranged in the system outer frame and below the conveying wire rope, and the sensing unit is arranged on the sensing support.

[0012] Further, the conveying module further comprises an adjusting unit, the adjusting unit comprises a tensioner and a tensioning motor, the tensioning motor is arranged on the conveying seat, the output end of the tensioning motor is connected with the tensioner, and the tensioner is in abutment with the conveying wire rope, so that when the tensioning motor is started, the tightness of the conveying wire rope is adjusted by adjusting the position of the tensioner.

[0013] Further, the screening module comprises a material distribution support, a material distribution electromagnetic valve, a mounting block and a material distribution pipe, the material distribution support is arranged in the system outer frame, the material distribution electromagnetic valve is arranged on the mounting block and is arranged on the material distribution support together, and the material distribution electromagnetic valve is connected with the material distribution pipe.

[0014] Further, the system further comprises at least two collecting units arranged at different positions, the collecting units are arranged at the end of the conveying wire rope, and when the material distribution electromagnetic valve is started, the connecting piece detected by the identification is sorted into one of the collecting units through the material distribution pipe.

[0015] Further, the collecting unit comprises a collecting box, a vibration motor, a supporting spring, a mounting seat and a mounting plate, the vibration motor is arranged at the lower end surface of the collecting box, the two ends of the supporting spring are connected with the lower end surface of the collecting box and the mounting seat respectively, and the mounting seat is connected with the mounting plate.

[0016] The application provides a connecting piece intelligent identification and screening system suitable for automatic assembly, which realizes accurate screening of connecting pieces through cooperative work of multiple function modules, and constructs a basic framework for system operation through an external module system frame and a control unit, so as to ensure physical integration and logical control of the function modules, the feeding module realizes continuous feeding of the connecting pieces through arrangement and conveying functions, and provides stable material flow for subsequent detection, the conveying module is designed in combination of a pair of conveying units and a conveying turntable, the posture of the connecting pieces is dynamically adjusted through rotating conveying wire ropes, so that the connecting pieces reach a positioning state suitable for detection in the moving process, the detection module implements identification and detection when conveying is synchronously performed, the boundary of the conveying wire rope is used as a spatial reference, and the relative position relationship between the connecting pieces and the wire rope is compared, so that slight defects such as surface scratches and corner defects of the rivet head can be accurately captured; and the screening module performs classification actions based on the detection results, so as to form a closed loop from detection to screening. Meanwhile, the conveying turntable drives the pair of wire ropes to form an adjustable clamping channel, the posture of the connecting pieces is automatically corrected in the conveying process, and the boundary of the wire rope is used to establish a detection coordinate system, so that the surface features of the connecting pieces and the reference line form a clear comparison relationship; the cooperative action of mechanical conveying and optical detection effectively solves the misjudgment problem caused by workpiece deviation in traditional static detection, and the identification accuracy of slight scratches and corner defects on the surface of the rivet head is significantly improved. BRIEF DESCRIPTION OF DRAWINGS

[0017] Figure 1 FIG. 1 is a structural schematic diagram of a screening system in embodiment 1 of the application; Figure 2 FIG. 2 is an internal structural schematic diagram of the screening system in embodiment 1 of the application; Figure 3 FIG. 3 is a structural schematic diagram of a feeding module in embodiment 1 of the application; Figure 4 FIG. 4 is a structural schematic diagram of a conveying module in embodiment 1 of the application; Figure 5 FIG. 5 is an internal overhead structural schematic diagram of the screening system in embodiment 1 of the application; Figure 6 FIG. 6 is a structural schematic diagram of a screening module in embodiment 1 of the application.

[0018] FIG. 6 is a structural schematic diagram of a screening module in embodiment 1 of the application. 1-external module, 2-feeding module, 3-detection module, 4-conveying module, 5-screening module, 6-connecting piece, 7-collection unit, 11-system frame, 12-control unit, 13-display unit, 14-interaction unit, 15-placing rack, 21-vibrating feeding unit, 22-linear feeding unit, 31-defect detection unit, 32-size detection unit, 33-sensor support, 311-defect detection support, 312-defect detection camera, 313-defect detection lens, 314-defect detection light source, 321-size detection support, 322-size detection camera, 323-size detection lens, 324-size detection light source, 41-conveying seat, 42-conveying unit, 43-conveying turntable, 44-conveying wire rope, 45-adjusting unit, 46-conveying motor, 451-tensioner, 452-tensioning motor, 51-distributing support, 52-distributing electromagnetic valve, 53-mounting block, 54-distributing pipe, 71-collection box, 72-vibration motor, 73-support spring, 74-mounting seat, 75-mounting plate, 76-falling material platform.

[0019] The purposes, functional features and advantages of the present application will be further described with reference to the embodiments and the accompanying drawings. DETAILED DESCRIPTION

[0020] The technical solutions in the embodiments of the present application will be clearly and completely described with reference to the accompanying drawings in the embodiments of the present application. Obviously, the described embodiments are only 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 work fall within the scope of protection of the present application.

[0021] It should be noted that all directional indications (such as up, down, left, right, front, back, etc.) in the embodiments of the present application are only used to explain the relative positional relationship, movement condition, etc. between components in a certain posture (as shown in the drawings), and if the certain posture changes, the directional indications will also change accordingly.

[0022] In the present application, unless otherwise explicitly specified and limited, the terms "connection", "fixation" and the like should be understood in a broad sense, for example, "fixation" can be fixed connection, or detachable connection, or integral; can be mechanical connection, or electrical connection; can be direct connection, or indirect connection through intermediate medium; can be internal connection of two elements or interaction relationship between two elements, unless otherwise explicitly limited. For those skilled in the art, the specific meaning of the above terms in the present application can be understood according to the specific circumstances.

[0023] In addition, if the description of "first", "second" and the like is involved in the embodiments of the present application, the description of "first", "second" and the like is only for the purpose of description, and cannot be understood as indicating or implying the relative importance of the indicated technical features or implicitly indicating the number of the indicated technical features. Therefore, the features limited by "first", "second" can be explicitly or implicitly included at least one of the features. In addition, the meaning of "and / or" appearing throughout the text includes three parallel schemes. For example, "A and / or B" includes A scheme, or B scheme, or A and B scheme. In addition, the technical solutions of each embodiment can be combined with each other, but it must be based on the realization of ordinary skilled in the art. When the combination of technical solutions appears contradictory or unachievable, it should be considered that the combination of technical solutions does not exist, and is not within the scope of protection required by the present application.

[0024] Embodiment 1: Please refer to the accompanying drawings Figures 1 to 6 The embodiment provides a connecting piece 6 intelligent identification screening system suitable for automatic assembly. The system is used for detecting and screening the connecting piece 6. The system comprises: An external module 1, which comprises a system outer frame 11, and a control unit 12 is arranged in the system outer frame 11; A feeding module 2, which is arranged in the system outer frame 11, and is used for completing the arrangement and conveying of the connecting piece 6; A conveying module 4, which is arranged in the system outer frame 11, and is used for completing the position adjustment of the connecting piece 6; A detection module 3, which is arranged in the system outer frame 11, and is used for completing the identification detection of the connecting piece 6 in the process of position adjustment of the connecting piece 6; A screening module 5, which is arranged in the system outer frame 11, and is used for completing the screening of the connecting piece 6 according to the identification detection result; The detection module 3 comprises a detection unit, the conveying module 4 comprises a conveying seat 41 and a pair of conveying units 42, the conveying unit 42 comprises a conveying turntable 43 and a conveying wire rope 44, the conveying turntable 43 is rotationally arranged on the conveying seat 41, the conveying wire rope 44 is movably arranged on the outer periphery of the conveying turntable 43, the connecting piece 6 is arranged between the two conveying wire ropes 44, and the position adjustment of the connecting piece 6 is completed through the rotation of the conveying turntable 43, and the detection unit is used for completing the identification detection of the connecting piece 6 by taking the boundary information of the conveying wire rope 44 as reference.

[0025] It should be noted that in the traditional existing connector 6 visual detection system, due to the lack of stable spatial reference in the dynamic conveying process, the surface defect recognition accuracy is limited; when the connector 6 is adjusted in posture in the conveying mechanism, the spatial coordinates of the connector 6 relative to the detection unit change continuously, and the traditional detection algorithm cannot effectively establish the mapping relationship between the dynamic coordinate system and the static detection reference. This technical defect directly leads to the fact that micro-defects with a surface scratch depth below 0.05 millimeters cannot be effectively captured, and abnormal shapes with an edge and corner loss area below 0.3 square millimeters are easily misjudged as normal tolerance range, causing the detection accuracy of key quality parameters to decrease to below 82%. The above problems directly affect the first-piece qualification rate of the automatic drilling and riveting system, causing the assembly line scrap rate to rise to 6.8%, and the single-batch rework time to increase by about 35 minutes.

[0026] It should also be noted that on an automatic drilling and riveting production line for an aircraft skin, alloy rivets with a diameter of 3 millimeters enter the flexible conveying belt at a speed of 2 pieces per second through a vibrating feeder, and when the conveying mechanism performs width self-adaptive adjustment, the rivets produce axial rotation and radial deviation in the double helical wire transmission. The traditional visual detection system cannot track the relative position of the rivet head and the conveying belt boundary in real time, resulting in a positioning deviation of the surface detection area exceeding ±0.5 millimeters.

[0027] Based on the above problems, the embodiment proposes a connector 6 intelligent recognition and screening system suitable for automatic assembly, which realizes accurate screening of the connector 6 through the cooperative work of multiple functional modules. Specifically, the system outer frame 11 and the control unit 12 of the external module 1 build the basic framework of system operation, ensuring the physical integration and logical control of each functional module. The feeding module 2 realizes continuous feeding of the connector 6 through arrangement and conveying functions, providing stable material flow for subsequent detection. The conveying module 4 adopts a combination design of paired conveying units 42 and conveying turntable 43, dynamically adjusts the posture of the connector 6 through rotating conveying wire 44, so that it reaches a suitable detection positioning state during movement. The detection module 3 implements recognition and detection when the conveying action is synchronized, uses the boundary of the conveying wire 44 as a spatial reference, and can accurately capture micro-defects such as rivet head surface scratches and edge and corner defects by comparing the relative position relationship between the connector 6 and the wire. The screening module 5 performs classification action based on the detection result, forming a complete detection-to-screening closed loop.

[0028] It can be understood that the core of the above process is the combination of the dynamic adjustment function of the flexible conveying mechanism and the spatial reference positioning of the visual detection system. The pair of wire ropes forms an adjustable clamping channel driven by the conveying turntable 43, automatically corrects the pose of the connecting piece 6 during the conveying process, and establishes a detection coordinate system with the wire rope boundary, so that the surface features of the connecting piece 6 and the reference line form a clear contrast relationship. The synergistic effect of mechanical conveying and optical detection effectively solves the misjudgment problem caused by the deviation of the workpiece in traditional static detection, and the recognition accuracy of the millimeter-level fine scratches and corner defects on the rivet head surface is significantly improved.

[0029] In some embodiments, the detection unit adopts a high-resolution camera and a special image processing algorithm, and takes the boundary information of the conveying wire rope 44 as a reference. When the connecting piece 6 moves with the conveying wire rope 44, the detection unit continuously captures the relative position relationship between the connecting piece 6 and the wire rope boundary. Therefore, the detection system can accurately position each area of the surface of the connecting piece 6, and can maintain high-precision detection even if the connecting piece 6 rotates or deviates.

[0030] In some embodiments, the image processing algorithm of the detection unit is optimized to be able to identify slight surface defects. By comparing the surface features of the connecting piece 6 with the preset standard model, the system can detect slight abnormalities such as surface scratches, depressions, and corner defects. Since the conveying wire rope 44 provides a dynamic spatial reference, the detection algorithm can more accurately determine the position and degree of the defect.

[0031] In some embodiments, the boundary information refers to the geometric feature data of the edge of the conveying wire rope 44, which can be realized by capturing the wire rope profile coordinates by a machine vision system, and establishing a spatial coordinate system as a detection reference.

[0032] In some embodiments, the synchronous position adjustment and identification detection refer to the timing cooperation of material conveying and quality detection, which can be realized by synchronously controlling the rotating speed of the conveying turntable 43 and the sampling frequency of the camera by using an encoder.

[0033] In the embodiment, the external module 1 further includes a display unit 13 and an interaction unit 14. The display unit 13 is used to display screening information, and the screening information includes the number and type of abnormalities of the connecting piece 6. The interaction unit 14 is used to complete information interaction.

[0034] It can be understood that the detection module 3 synchronously completes the identification detection of the connecting piece 6 during the position adjustment of the conveying wire rope 44, and the control unit 12 transmits the detection result to the display unit 13. Among them, the display unit 13 presents the abnormal number cumulative value and the abnormal type proportion pie chart in a visual form, for example, after continuously detecting 200 connecting pieces 6, the display interface updates to display that the current abnormal number is 15, among which the surface scratch accounts for 60%, the size deviation accounts for 30%, and the identification error accounts for 10%; the operator judges whether the quality fluctuation of the current batch of connecting pieces 6 exceeds the preset range by observing the abnormal type distribution trend.

[0035] It can also be understood that when the number of abnormalities exceeds the set threshold, the interactive unit 14 receives the external input pause instruction, the control unit 12 immediately interrupts the operation of the conveying module 4, and the operator adjusts the triggering condition of the distribution electromagnetic valve 52 through the interactive unit 14, for example, the screening level of the surface scratch type abnormality is improved from level two to level one; the adjusted parameters are synchronized to the screening module 5 through the control unit 12, and the distribution pipe 54 sorts the defective connecting pieces 6 to a specific collection box 71. The detection report can be exported through the interactive unit 14, including the timestamp, the abnormal type code and the processing personnel operation record, forming a complete traceability chain.

[0036] Through the above, the embodiment realizes information visualization and real-time interaction in the screening process of the connecting piece 6, and the display unit 13 intuitively presents the abnormal number and type, so that the operator can timely master the screening state and quickly identify quality problems; the interactive unit 14 provides a two-way information transmission channel, enhancing the operability and adaptability of the system. Not only improves the monitoring efficiency of the screening process, but also provides data support for production management decision-making. Through real-time data display and interaction function, the operator can quickly adjust the screening parameters or troubleshoot equipment problems, thereby improving the production efficiency and quality control ability. In addition, the transparency and traceability of information have been significantly improved, which is beneficial to subsequent quality analysis and problem tracing.

[0037] In the embodiment, the system outer frame 11 is provided with a placing rack 15, the feeding module 2 includes a vibration feeding unit 21 and a linear feeding unit 22, the vibration feeding unit 21 and the linear feeding unit 22 are both arranged on the placing rack 15, and the horizontal height of the input end of the linear feeding unit 22 is higher than that of the output end.

[0038] It can be understood that after the connecting piece 6 is arranged in a direction by the vibration feeding unit 21, it slides downward along the inclined track under the action of gravity, and when the vibration feeding unit 21 vibrates at a certain frequency, the connecting piece 6 is pushed to the input end of the linear feeding unit 22. Due to the height difference between the input end and the output end, the connecting piece 6 moves at a constant speed along the inclined track under the action of gravity.

[0039] In some embodiments, when the inclination angle is 8°, the average conveying speed of the connecting piece 6 can reach 0.5 m / s without additional power driving. During the conveying process, the wear-resistant coating on the track surface can reduce the friction coefficient to below 0.15, avoiding scratches on the surface of the connecting piece 6. The rigid support structure of the placing rack 15 can suppress vibration transmission, so that the vibration amplitude difference between the linear feeding unit 22 and the vibrating feeding unit 21 is controlled within ±0.1 mm, ensuring that there is no material accumulation at the junction of the two. By adjusting the matching relationship between the inclination angle and the vibration frequency, the spacing of the connecting piece 6 during the conveying process can be stably maintained within 2-3 times the diameter range, providing a uniformly distributed material flow for the subsequent detection module 3.

[0040] Based on the above, stable support and efficient conveying of the feeding module 2 are achieved. The placing rack 15 provides an integrated installation basis for the vibrating feeding unit 21 and the linear feeding unit 22, ensuring that their positions are fixed and stable in cooperation within the system outer frame 11, avoiding displacement deviation caused by mechanical vibration. The linear feeding unit 22 adopts an inclination design with the input end higher than the output end, utilizing gravity to assist the natural sliding of the connecting piece 6 from the input end to the output end, reducing stagnation or accumulation caused by insufficient external driving. Not only does this reduce energy consumption, but it also maintains a continuous and uniform flow state of the connecting piece 6 during the conveying process through gravity guidance, thereby improving the reliability and efficiency of the entire feeding module 2 and laying a foundation for the precise operation of the subsequent detection and screening module 5.

[0041] In this embodiment, the detection module 3 includes a defect detection unit 31 arranged above the conveying wire rope 44. The defect detection unit 31 includes a defect detection bracket 311, a defect detection camera 312, a defect detection lens 313, and a defect detection light source 314. The defect detection bracket 311 is arranged in the system outer frame 11. The defect detection camera 312 and the defect detection lens 313 are arranged on the defect detection bracket 311 in cooperation. The defect detection light source 314 is arranged on the defect detection bracket 311 and placed in front of the defect detection lens 313.

[0042] It can be understood that when the conveying wire rope 44 carries the connecting piece 6 through the detection area, the ring-shaped light source projects light with a normal illumination angle of 0° to 15°, for example, with an inclination angle of 8°, forming a uniform light field without shadows on the surface of the rivet head. The camera synchronously captures images at a capture rate of 30 frames per second. The lens is configured with a combination of 5 times optical magnification and 2 million pixel resolution, which can clearly present millimeter-level surface scratches. The rigid fixed structure of the bracket can suppress image blur caused by equipment vibration. In addition, the front layout of the light source makes the light directly cover the surface of the measured object, which effectively enhances the gray difference of the edge and corner defects compared with the lateral lighting scheme.

[0043] Based on the above, high-precision defect detection in the dynamic conveying state of the connecting piece 6 is realized. Specifically, the defect detection unit 31 is arranged above the conveying wire rope 44, and the defect detection camera 312 cooperates with the lens to enable high-resolution image acquisition of the surface of the connecting piece 6. The defect detection light source 314 is placed at the front end of the lens to form uniform front lighting in the detection area, effectively eliminating shadow interference and enhancing the contrast of surface micro-defects. This arrangement enables precise capture of micro-abnormalities such as scratches and corner loss on the rivet head surface during dynamic conveying.

[0044] In this embodiment, the detection module 3 further includes a size detection unit 32 arranged to the side of the conveying wire rope 44. The size detection unit 32 includes a size detection bracket 321, a size detection camera 322, a size detection lens 323, and a size detection light source 324. The size detection bracket 321 is arranged in the system outer frame 11. The size detection camera 322 and the size detection lens 323 are arranged on the size detection bracket 321 after cooperation. The size detection light source 324 is arranged in front of the size detection lens 323 to enable the conveying wire rope 44 to be located between the size detection lens 323 and the size detection light source 324.

[0045] It can be understood that when the connecting piece 6 moves with the conveying wire rope 44 to the detection station, the size detection light source 324 projects a parallel light beam forward, enabling the connecting piece 6 to form a high-contrast silhouette image in the size detection lens 323. The size detection camera 322 continuously acquires images at a rate of 30 frames per second, and extracts the contour feature points of the connecting piece 6 through an edge detection algorithm. The length of the connecting piece 6 can be obtained by calculating the distance between the first and last feature points, and the diameter can be determined by the maximum width value of the contour. During detection, the rigid structure of the size detection bracket 321 can suppress image blur caused by equipment vibration. After comparing the detection data with the preset standard parameters, the distribution electromagnetic valve 52 triggers the corresponding action channel according to the deviation threshold to sort the out-of-tolerance connecting piece 6 to the independent collection unit 7.

[0046] In this embodiment, the detection module 3 further includes a sensing bracket 33 and a sensing unit. The sensing bracket 33 is arranged in the system outer frame 11 and below the conveying wire rope 44. The sensing unit is arranged on the sensing bracket 33.

[0047] It should be noted that when the conveying turntable 43 drives the conveying wire rope 44 to move, the connecting piece 6 slides or deviates between the two conveying wire ropes 44, at which time the sensing unit collects the vibration amplitude or position deviation of the conveying wire rope 44 in real time, for example, when the displacement sensor detects that the conveying wire rope 44 deviates laterally by more than 0.1 mm, the data is fed back to the control unit 12. The control unit 12 dynamically corrects the visual detection reference line of the detection unit according to the displacement data, for example, adjusts the original boundary reference line of the conveying wire rope 44 by 0.08 mm, so that the defect detection unit 31 can identify the surface scratches or corner missing of the rivet head when shooting the image of the connecting piece 6 based on the corrected reference line. At the same time, when the pressure sensor detects that the contact pressure between the conveying wire rope 44 and the connecting piece 6 is lower than the preset threshold, it indicates that the connecting piece 6 has the risk of disengagement, at which time the control unit 12 triggers an alarm signal and suspends the operation of the conveying module 4, avoiding detection failure caused by abnormal position of the connecting piece 6.

[0048] In the embodiment, the conveying module 4 further comprises an adjusting unit 45, which comprises a tensioner 451 and a tensioning motor 452, the tensioning motor 452 is arranged on the conveying seat 41, the output end of the tensioning motor 452 is connected with the tensioner 451, and the tensioner 451 abuts against the conveying wire rope 44, so that when the tensioning motor 452 is started, the tightness of the conveying wire rope 44 is adjusted by adjusting the position of the tensioner 451.

[0049] It can be understood that when the conveying wire rope 44 is relaxed due to wear or load change, the tensioning motor 452 receives the adjusting signal sent by the control unit 12, drives the threaded rod to move the tensioner 451 along the slide rail. The tensioner 451 contacts the conveying wire rope 44 in the form of a pulley or a pressing block, and adjusts the overall tension by changing the position of the contact point to make the conveying wire rope 44 elastically deform.

[0050] Specifically, the tensioner 451 can increase the tension of the conveying wire rope 44 by 3-5 N per millimeter of movement, and the adjustment frequency can be more than 10 times per second. During the continuous rotation of the conveying turntable 43, the tensioner 451 compensates for the elongation of the conveying wire rope 44 in real time, and controls the tension fluctuation within ±2%. By maintaining the constant tension of the conveying wire rope 44, the relative position accuracy of the connecting piece 6 to the detection module 3 is maintained during the conveying process. When different models of connecting pieces 6 are replaced, the tensioning motor 452 automatically adjusts the position of the tensioner 451 according to the preset parameters, so that the spacing of the conveying wire rope 44 adapts to the size of the connecting piece 6, thereby effectively reducing the switching time.

[0051] In the embodiment, the screening module 5 comprises a distribution support 51, a distribution electromagnetic valve 52, a mounting block 53 and a distribution pipe 54. The distribution support 51 is arranged in the system outer frame 11. The distribution electromagnetic valve 52 is arranged on the mounting block 53 and is jointly arranged on the distribution support 51. The distribution electromagnetic valve 52 is connected with the distribution pipe 54.

[0052] It can be understood that the distribution support 51 is fixed on the inner frame of the system outer frame 11 by a rigid connection mode as a basic support structure to form a stable mounting platform. The mounting block 53 is formed by machining a mounting groove matched with the shape of the distribution electromagnetic valve 52. After the electromagnetic valve is embedded in the mounting groove, it is fixed by a pressing plate to ensure that it does not displace in a vibrating environment. The distribution pipe 54 is butted with the outlet end of the electromagnetic valve through a flange. The flange bolts are locked in three steps by a diagonal tightening mode to ensure that the connecting surface is uniformly stressed. When the detection module 3 identifies an unqualified connecting piece 6, the control unit 12 triggers the distribution electromagnetic valve 52 to act. The valve core completes the switching between the open and closed states within 20 to 50 milliseconds. The distribution pipe 54 guides the connecting piece 6 to a specified collection unit 7 according to the instruction.

[0053] In the embodiment, the system further comprises at least two collection units 7 arranged at different positions. The collection units 7 are arranged at the end of the conveying wire rope 44. When the distribution electromagnetic valve 52 is started, the distribution pipe 54 sorts the identified connecting piece 6 into one of the collection units 7.

[0054] In the embodiment, the collection unit 7 comprises a collection box 71, a vibrating motor 72, a supporting spring 73, a mounting seat 74 and a mounting plate 75. The vibrating motor 72 is arranged at the lower end surface of the collection box 71. The two ends of the supporting spring 73 are respectively connected with the lower end surface of the collection box 71 and the mounting seat 74. The mounting seat 74 is connected with the mounting plate 75.

[0055] When the detection module 3 identifies that the connecting piece 6 has a defect, the distribution electromagnetic valve 52 receives a control signal and switches the conduction direction of the distribution pipe 54 to guide the defective piece into the specified collection unit 7. When the collection box 71 receives the sorted connecting piece 6, the vibrating motor 72 is started and generates horizontal or vertical vibration to make the connecting piece 6 uniformly distributed in the collection box 71 to avoid jamming caused by accumulation. The supporting spring 73 elastically deforms when the collection box 71 is impacted by vibration to absorb mechanical energy and reduce the vibration amplitude transmitted to the mounting seat 74. The rigid connection structure of the mounting seat 74 and the mounting plate 75 can inhibit the displacement deviation of the collection box 71 in the vibration process to ensure the stability of the sorting path. Through the combined action of the directional switching of the distribution path and the vibration buffering, the qualified pieces and the defective pieces are physically isolated and stored. At the same time, the damage probability of the surface of the connecting piece 6 caused by collision is effectively reduced.

[0056] Embodiment 2: As a preferred embodiment, the specific implementation of this embodiment is as follows: The system frame 11 is made of aluminum alloy profile, with the characteristics of light weight and high strength. The control unit 12 includes an industrial computer and a programmable logic controller, installed in the control cabinet inside the system frame 11. The feeding module 2 uses a combination of vibration disc and linear feeder to ensure the orderly arrangement and stable delivery of the connecting piece 6.

[0057] The conveying seat 41 of the conveying module 4 is processed from stainless steel plate, and the surface is treated for corrosion resistance. The conveying turntable 43 is supported by high-precision bearings and driven by a servo motor, which can realize precise angle control. The conveying wire rope 44 is made of wear-resistant nylon material, with good flexibility and durability. The spacing between the two conveying wire ropes 44 can be adjusted by the adjusting mechanism to adapt to different specifications of the connecting piece 6.

[0058] The detection unit of the detection module 3 includes an industrial camera, a lens, and a light source system. The camera uses a high-resolution CMOS sensor and is equipped with a macro lens to capture the small details on the surface of the connecting piece 6. The light source system uses an adjustable angle LED array to provide uniform and controllable lighting conditions. The detection unit is fixed above the conveying wire rope 44 by a precision bracket, maintaining the optimal imaging distance and angle.

[0059] The image processing algorithm is developed based on deep learning technology and can identify various types of connecting piece 6 defects after a large number of sample training. The algorithm processes image data in real time and compares the surface features of the connecting piece 6 with the preset qualified standard to determine whether there is an abnormality.

[0060] The screening module 5 uses a high-speed pneumatic sorting device to control the airflow direction according to the detection results, guiding the unqualified connecting piece 6 to a separate collection container. The qualified connecting piece 6 continues to be transmitted along the conveying wire rope 44 to the next process.

[0061] When the system is running, the connecting piece 6 enters the conveying module 4 from the feeding module 2 and is placed between the two conveying wire ropes 44. The wire rope is moved by the conveying turntable 43, making the connecting piece 6 pass under the detection unit at a constant speed. The detection unit continuously collects images and analyzes them with the edge of the conveying wire rope 44 as a reference. Once a defect is detected, the system immediately sends a signal to the screening module 5 to trigger the sorting action. The entire process is highly automated and can achieve continuous and efficient screening of the connecting piece 6.

[0062] By the above scheme, the present application can effectively solve the problem of difficult identification of abnormal conditions such as small scratches on the rivet head surface, corner loss, etc. in the prior art. Since the conveying wire rope 44 is used as a dynamic space reference, the system can maintain high-precision detection capability during the movement of the connecting piece 6. This method significantly improves the identification rate of micro-defects, especially for surface scratches and corner loss and other subtle abnormalities. At the same time, the rotating function of the conveying turntable 43 enables the connecting piece 6 to present multiple angles during the detection process, ensuring omnidirectional defect inspection. This dynamic detection method not only improves the accuracy of detection, but also maintains the efficient operation of the system, meeting the production rhythm requirements of the automated assembly line. In addition, the real-time screening function effectively prevents unqualified connecting pieces 6 from entering the subsequent assembly process, thereby improving the overall product quality and reducing rework and quality problems caused by defective connecting pieces 6.

[0063] Embodiment 3: To make the technical solutions of the present application clearer, the specific algorithm for size detection is explained in detail here. In this embodiment, the connecting piece 6 in the conveying process is not considered as a collection of a series of static images, but its movement and posture change is considered as a useful information dimension. Through spatio-temporal data fusion and active optical perception, passive detection is upgraded to dynamic three-dimensional reconstruction and analysis, thereby achieving a leap in detection accuracy and robustness. Specifically: Unlike the traditional single-frame image measurement mode, by taking advantage of the uniform linear motion characteristics of the connecting piece 6 driven by the conveying wire rope 44, this embodiment reconstructs a one-dimensional projection profile of the connecting piece 6 with ultra-high resolution and no motion distortion by time series sampling of the data in a specific area of the image sensor, thereby achieving millimeter-level, even micrometer-level precision size measurement.

[0064] The specific steps of size detection include: Linear array mode sampling: The side size detection camera no longer collects images in full frame, but is virtually converted into a linear array camera, and only one or several columns of pixels (defined as a sampling line) from the center of the camera sensor are continuously read at a very high frequency (kHz level); Dynamic reference calibration: Before the connecting piece 6 enters the sampling line, the edge of the conveying wire rope is first captured and tracked as it passes through the sampling line. Since the wire rope diameter is known and can be considered constant, the time it takes for the wire rope edge to pass through the sampling line , combined with the accurate speed v provided by the conveying motor encoder, can be used to calibrate the pixel equivalent K in real time and monitor the stability of the conveying speed.

[0065] Space-time data acquisition: When the connector 6 body starts to pass through the sampling line, the system records the time point when each row of pixels changes from light (background) to dark (connector) and from dark to light. This process continues until the entire connector passes through the sampling line.

[0066] One-dimensional profile reconstruction: Convert time series data into spatial data. The width of the connector at each pixel point on the sampling line (representing a certain position in the diameter direction) is the duration of time that the pixel point is blocked and the conveying speed and the conveying speed v determined.

[0067] Size calculation and determination: Through the reconstructed complete one-dimensional profile data array, the total length and diameter of the connector at any position can be calculated extremely accurately. Analyze the envelope and maximum value of the profile, compare it with the pre-set standard model, and determine whether the size is qualified.

[0068] Some of the expressions involved in the above process are as follows: For real-time dynamic calibration and speed verification: ; Where, v is the real-time calculated conveying speed of the connector; is the standard diameter of the conveying wire rope 44 (known); is the time required for the conveying wire rope 44 to completely pass through a certain pixel point on the sampling line. By comparing this v value with the motor encoder reading, the stability of the system operation can be double-verified.

[0069] For space-time profile reconstruction: ; Where, is the actual width (mm) of the connector 6 at the radial position y i ; v is the conveying speed; is the duration (s) that the pixel point y i is blocked by the connector 6.

[0070] It can be understood that for any one pixel point y i on the sampling line (representing a position of the connector 6 in the radial direction), the corresponding part of the width of the connector 6 satisfies the above expression. Through all the blocked pixel points y iThe corresponding actual width is calculated, and a set of discrete data points is obtained This set of data accurately describes the axial profile of the connector.

[0071] For size calculation, Total length The overall envelope length of the reconstructed profile satisfies: ; Total diameter The number of pixel rows covered by the reconstructed profile multiplied by the physical size of the pixel satisfies: ; Wherein, are the indices of the highest and lowest pixel rows blocked by the connected component, respectively; is the physical size of a single pixel of the camera (mm).

[0072] It can be understood that by converting time accuracy into spatial accuracy, measurement accuracy far exceeding the original resolution of the camera can be easily achieved, completely immune to motion blur, and through dynamic calibration, the small fluctuations of the conveying speed have strong robustness.

[0073] Example 4: In order to make the technical solutions of the present application more clear, the specific algorithm for defect detection is explained in detail here, specifically: combining active multi-directional illumination, passive two-dimensional image defect detection is upgraded to active three-dimensional surface normal vector field reconstruction and analysis, which can find micron-level depth scratches, pits and surface ripples that are difficult to identify by the human eye and traditional 2D vision.

[0074] Defect detection specifically includes the following steps: Active illumination and synchronous acquisition: the defect detection light source is modified into a partitioned high-speed stroboscopic LED array light source (such as four partitions, representing incident light in the upper, lower, left and right four directions, respectively); when the connector head passes through the detection area, the system stroboscopically flashes the light sources in different directions in a very short time (milliseconds), and synchronously captures 3-4 images; based on the stable conveying brought by the conveying module 4, it can be considered that these images are taken under the condition of the same position and only different light directions.

[0075] Normal vector field calculation: using photometric stereo algorithm, according to these images under different illumination, the three-dimensional surface normal vector of each pixel point on the head surface of the connector 6 is solved N This normal vector field contains the accurate three-dimensional geometric information of the object surface.

[0076] Reference normal vector field establishment: the system pre-models a standard, defect-free connector 6, and generates an ideal reference normal vector fieldN ideal .

[0077] Differential field analysis: compare the real-time calculated normal vector field N actual with the reference normal vector field N ideal A smooth, defect-free surface has a normal vector field that is smooth and continuously changing, while a scratch or a pit, by its nature, is a sharp change in the surface orientation, which will result in a sharp, non-continuous abnormal vector in the normal vector field.

[0078] Defect quantification and determination: quantify the severity of the defect by calculating the angular deviation between the real normal vector field and the reference normal vector field, or calculating the local curvature / gradient of the real normal vector field itself, for example, a scratch will show as a line of abnormal gradient, while a pit will show as an area of abnormal gradient.

[0079] Solution of photometric stereo normal vector: Based on the Lambertian reflection model, the pixel intensity I is related to the surface normal vector N , the light source direction vector L by the expression: ; where I is the gray value of the pixel; ρ is the surface reflectivity; N is the unit normal vector of the surface corresponding to the pixel, and satisfies , N x ，N y ，N z represent the components in the three directions, respectively; L is the unitized light source direction vector.

[0080] When there are 3 light sources L1L2L3 in different directions and corresponding 3 images I1I2I3, a linear equation system can be constructed: ; Let the matrix be L and the vector be ; then ; Solve to get the vector g. Since N is a unit vector, the surface reflectivity satisfies: ; The normal vector satisfies: ; Thus, the three-dimensional normal vector of each pixel point is calculated.

[0081] For quantitative determination of defect detection, the following two methods are preferred: Method 1: An angle deviation method is adopted, that is, the angle deviation between the actual normal vector of each pixel and the ideal normal vector is calculated , and satisfies: ; And an angle threshold is set, when the angle deviation is greater than the angle threshold, the point is defined as a defect point, and is marked as an abnormal connecting piece that needs to be screened.

[0082] Method 2: A local gradient method is adopted, that is, the gradients of the normal vector field in the x and y directions are calculated, and the modulus is calculated, and satisfies: Wherein, ∇ represents the Laplace operator.

[0083] It can be understood that the gradient value of a smooth surface is low, and defects such as scratches and pits will cause the gradient value to rise sharply, and by setting a gradient threshold, when the local gradient modulus is greater than the gradient threshold, it is defined as a defect, and the corresponding connecting piece 6 is marked as an abnormal connecting piece that needs to be screened.

[0084] The above method jumps from two-dimensional gray scale analysis to three-dimensional geometric analysis, and can consider depth information. It is not sensitive to light changes, and can stably detect same-color scratches, shallow pressure injuries and small pits that are easily missed in traditional methods due to insufficient contrast, and realizes super-precision detection of the surface micro-topography.

[0085] The above is only a preferred embodiment of the present application, and does not limit the patent scope of the present application, and any equivalent structure or equivalent process transformation using the content of the specification and drawings, or direct or indirect application in other related technical fields, are also included in the patent protection scope of the present application.

Claims

1. A connector intelligent identification screening system suitable for automated assembly, characterized in that, The system is used for detecting and screening the connector, and comprises: An external module, which comprises a system outer frame, and a control unit is arranged in the system outer frame; A feeding module, which is arranged in the system outer frame and is used for completing the arrangement and conveying of the connector; A conveying module, which is arranged in the system outer frame and is used for completing the position adjustment of the connector; A detection module, which is arranged in the system outer frame and is used for completing the identification detection of the connector during the position adjustment of the connector; A screening module, which is arranged in the system outer frame and is used for completing the screening of the connector according to the identification detection result; The detection module comprises a detection unit, the conveying module comprises a conveying seat and a pair of conveying units, the conveying unit comprises a conveying turntable and a conveying wire rope, the conveying turntable is rotationally arranged on the conveying seat, the conveying wire rope is movably arranged on the outer periphery of the conveying turntable, the connector is arranged between the two conveying wire ropes, and the position adjustment of the connector is completed through the rotation of the conveying turntable, and the detection unit is used for completing the identification detection of the connector by taking the boundary information of the conveying wire rope as the reference.

2. The connector intelligent identification and screening system suitable for automated assembly of claim 1, wherein, The external module further comprises a display unit and an interaction unit, the display unit is used for displaying screening information, the screening information comprises the abnormal quantity and the abnormal type of the connector, and the interaction unit is used for completing the information interaction.

3. The connector intelligent identification and screening system suitable for automated assembly of claim 1, wherein, The system outer frame is provided with a placing rack, the feeding module comprises a vibration feeding unit and a linear feeding unit, the vibration feeding unit and the linear feeding unit are arranged on the placing rack, and the horizontal height of the input end of the linear feeding unit is higher than that of the output end.

4. The connector intelligent identification and screening system suitable for automated assembly of claim 1, wherein, The detection module comprises a defect detection unit, the defect detection unit is arranged above the conveying wire rope, the defect detection unit comprises a defect detection support, a defect detection camera, a defect detection lens and a defect detection light source, the defect detection support is arranged in the system outer frame, the defect detection camera and the defect detection lens are arranged on the defect detection support after being matched, and the defect detection light source is arranged on the defect detection support and located at the front end of the defect detection lens.

5. The connector intelligent identification and screening system suitable for automated assembly of claim 1, wherein, The detection module further comprises a size detection unit, the size detection unit is arranged on the side of the conveying wire rope, the size detection unit comprises a size detection support, a size detection camera, a size detection lens and a size detection light source, the size detection support is arranged in the system outer frame, the size detection camera and the size detection lens are arranged on the size detection support after being matched, and the size detection light source is arranged in front of the size detection lens so that the conveying wire rope is located between the size detection lens and the size detection light source.

6. The connector intelligent identification and screening system suitable for automated assembly of claim 1 or 5, wherein, The detection module further comprises a sensing support and a sensing unit, the sensing support is arranged in the system outer frame and below the conveying wire rope, and the sensing unit is arranged on the sensing support.

7. The connector intelligent identification and screening system suitable for automated assembly of claim 1, wherein, The conveying module further comprises an adjusting unit, the adjusting unit comprises a tensioner and a tensioning motor, the tensioning motor is arranged on the conveying seat, the output end of the tensioning motor is connected with the tensioner, and the tensioner is in abutment with the conveying wire rope, so that when the tensioning motor is started, the tightness of the conveying wire rope is adjusted by adjusting the position of the tensioner.

8. The connector intelligent identification and screening system suitable for automated assembly of claim 1, wherein, The screening module comprises a distribution support, a distribution electromagnetic valve, a mounting block and a distribution pipe, the distribution support is arranged in the system outer frame, the distribution electromagnetic valve is arranged on the mounting block and is jointly arranged on the distribution support, and the distribution electromagnetic valve is connected with the distribution pipe.

9. The connector intelligent identification and screening system suitable for automated assembly of claim 8, wherein, The system further comprises at least two collecting units arranged at different positions, the collecting units are arranged at the end of the conveying wire rope, and when the distribution electromagnetic valve is started, the connecting piece detected by the identification is sorted into one of the collecting units through the distribution pipe.

10. The connector intelligent identification and screening system suitable for automated assembly of claim 9, wherein, The collecting unit comprises a collecting box, a vibration motor, a supporting spring, a mounting seat and a mounting plate, the vibration motor is arranged on the lower end surface of the collecting box, the two ends of the supporting spring are connected with the lower end surface of the collecting box and the mounting seat respectively, and the mounting seat is connected with the mounting plate.

Images