A multi-unit self-tapping screw automatic locking production line

Abstract

This invention discloses a multi-unit automatic self-tapping screw fastening production line, comprising a frame, a circulating conveying system, a feeding unit, a CCD vision positioning unit, and at least one fastening unit. The fastening unit includes a fastening drive mechanism and a fastening execution mechanism. The fastening execution mechanism includes a guide preload mechanism, an axial floating mechanism, and a bit assembly. The guide preload mechanism is used for screw centering and workpiece preload. The axial floating mechanism employs a nonlinear variable stiffness elastomer to dynamically adjust the axial feed force. The control system controls the motion path based on CCD vision positioning, synchronously collects torque and displacement data, constructs the fastening quality characteristic parameter K = dT / dS, and determines the fastening state in real time through a torque-displacement weighted fusion index Q, achieving stripped threads and floating lock identification. The production line also includes multi-unit dynamic task scheduling, bit wear detection and life prediction, adaptive parameter adjustment, and a remote monitoring and traceability platform. This invention improves fastening quality and production efficiency.

Description

Technical Field

[0001] This invention relates to the field of self-tapping screws, and in particular to an automatic fastening production line for multi-unit self-tapping screws. Background Technology

[0002] Currently, in large-scale manufacturing sectors such as electronics, automotive parts, and home appliances, threaded connections are one of the most common assembly methods. Self-tapping screws are widely used due to their advantages such as not requiring pre-tapping and reliable connection. With the continuous improvement of industrial automation, automatic fastening equipment is gradually replacing traditional manual fastening operations, effectively improving production efficiency and assembly quality consistency.

[0003] In existing technologies, automatic screw fastening equipment typically includes basic components such as a feeding unit, a motion mechanism, a screw fastening actuator, and a control system. For example, patent application CN117206888A discloses an automatic screw fastening machine with torque detection and waste material recycling. This equipment drives the electric screwdriver assembly to move through a three-axis motion mechanism and is equipped with an electric screwdriver torque detector to detect torque before fastening. Simultaneously, it collects abnormal screws through a waste material collection box. This solution improves the automation level of the equipment to some extent, but its screw fastening actuator only uses a simple spring buffer structure, making it difficult to accurately adjust the feed force according to different working conditions during the fastening process. Furthermore, it lacks the ability to jointly detect torque and displacement during screw fastening, and cannot effectively identify common fastening defects such as stripped threads and loose screws.

[0004] Patent CN222552737U further proposes a screw fastening device. This device has a camera fixedly installed on the screw fastening machine body to monitor the degree of sway after the screw is attracted in real time. When excessive sway is detected, a screw throwing operation is performed. Although this solution can avoid the degradation of fastening quality caused by screw sway, its visual inspection function is limited to the screw posture detection before fastening, lacking the ability to accurately position the workpiece hole. Moreover, it still relies on the traditional open-loop control method during the fastening process, making it difficult to adapt to changes in process parameters for different workpieces and screw specifications.

[0005] Based on the above-mentioned publicly available documents and existing technology, the following deficiencies still exist; First, there is a lack of effective screw alignment and preload mechanisms. The existing equipment's locking actuator lacks radial guidance and axial preload functions when the screw is aligned with the workpiece hole, which makes the screw prone to deflection, leading to quality problems such as stripped threads and misaligned screws during the locking process.

[0006] Second, the mechanical adaptability of the locking process is poor. Existing equipment mostly uses elastic buffer elements with fixed stiffness, which cannot dynamically adjust the axial feed characteristics according to the locking process. Impacts are easily generated in the initial stage of locking, and it is difficult to ensure accurate feed in the final stage of locking, which affects the locking quality.

[0007] Third, the methods for detecting locking quality are limited. Most existing equipment relies solely on peak torque to determine if locking is complete, lacking the ability to jointly analyze torque and displacement. This makes it difficult to accurately identify abnormal states such as slippage and floating locks, and also makes it impossible to achieve full data traceability of the locking process.

[0008] Fourth, the equipment is not intelligent enough. Most of the existing equipment is single-station operation or simple parallel connection, lacking dynamic task scheduling capabilities and intelligent functions such as adaptive parameter adjustment, bit wear prediction, and equipment comprehensive efficiency trend analysis, making it difficult to adapt to the flexible production needs of multiple varieties and variable batches.

[0009] Fifth, the production line layout and material circulation efficiency need to be improved. Most of the existing equipment adopts a single-layer conveyor belt or manual loading and unloading method, which makes it difficult to realize the automatic circulation processing of workpieces between the upper and lower conveyor belts, thus limiting the continuous operation capability and space utilization of the production line. Summary of the Invention

[0010] One object of the present invention is to provide a multi-unit self-tapping screw automatic fastening production line that at least solves any of the above-mentioned technical problems.

[0011] In particular, the present invention provides a multi-unit automatic fastening production line for self-tapping screws, including a frame, a conveyor line, a feeding unit, a CCD vision positioning unit and at least one fastening unit; The fastening unit includes a fastening drive mechanism and a fastening execution mechanism. The fastening execution mechanism includes a guide preload mechanism, which is used to radially guide the self-tapping screw and preload and position the workpiece before fastening. An axial floating mechanism is disposed between the locking drive mechanism and the bit assembly.

[0012] Furthermore, both the locking and fastening actuator and the CCD vision positioning unit are mounted on the frame and located above the first conveyor belt. The CCD vision positioning unit is located upstream of the locking and fastening actuator and is used to first locate the mounting holes on the mounting plate and record the mounting plate number, transmit the positioning data and temporarily store it in the locking and fastening actuator, and perform locking when the mounting plate with the corresponding number moves to the locking and fastening actuator.

[0013] Furthermore, the locking mechanism includes a feeding unit and a feeding hopper. The feeding unit is connected to the feeding hopper, and a support plate, a first transverse slide rail, a first longitudinal slide block, and a first longitudinal slide rail are arranged below the feeding hopper. The locking mechanism is connected to a first fixed plate through the support plate. The first fixed plate is fixedly connected to the first longitudinal slide block. The first longitudinal slide block is slidably arranged with the first longitudinal slide rail and is also moved along the length of the support plate through the first transverse slide rail. The CCD vision positioning unit controls the movement through the same structure as the locking mechanism, specifically including a second transverse slide rail, a second longitudinal slide block, a second longitudinal slide rail, and a second transverse fixed plate for fixation, so as to achieve independent motion control.

[0014] Furthermore, the frame is also equipped with a control panel, which is electrically connected to the control system and is used to display the production line operating status, fastening quality data and equipment fault information, and to accept parameter configuration input from the operator.

[0015] Furthermore, the first mounting plate placement box and the second mounting plate placement box are respectively provided with a driving mechanism. The driving mechanism is connected to the conveying mechanism and is used to drive the conveying mechanism to move up and down along the vertical rail, so as to realize the cyclic switching of the mounting plate between the first conveyor belt and the second conveyor belt.

[0016] Furthermore, the production line also includes a control system. This control system controls the movement path of the fastening unit based on the workpiece position information acquired by the CCD vision positioning unit, and simultaneously collects torque data and axial displacement data during the fastening process to determine the fastening status of the self-tapping screws. The control system is configured to construct a fastening quality characteristic parameter K based on the ratio of the torque change rate to the axial displacement change rate. Where T is the real-time torque, S is the axial displacement, and t is time; and through a preset threshold range Real-time classification and determination of the lock status.

[0017] Furthermore, the axial floating mechanism includes a floating sleeve, an elastic damping element disposed within the floating sleeve, and an axial displacement sensor; the elastic damping element is a nonlinear variable stiffness elastic body, whose stiffness changes continuously with the compression amount, and the stiffness function k(x) is expressed as: Where x is the compression amount, k0 is the initial stiffness, k1 is the stiffness coefficient, and n is the nonlinear exponent, where n>1; the stiffness curve is configured to provide low stiffness buffering in the initial stage of locking and high stiffness support in the final stage of locking; the axial displacement sensor detects the axial displacement of the bit assembly in real time and feeds it back to the control system; the control system dynamically adjusts the axial feed speed of the locking drive mechanism according to the deviation between the real-time axial displacement and the preset displacement curve, so that the axial displacement tracking error is minimized. minimize.

[0018] Furthermore, the control system performs joint torque and displacement detection and constructs a locking quality judgment model; the model evaluates the locking state by calculating a weighted fusion index of the differential value of real-time torque relative to angle and the differential value of axial displacement relative to angle; the weighted fusion index is defined as: Where T is the real-time torque, Let S be the screw rotation angle, and S be the axial displacement. and The preset weighting coefficients are used, and they satisfy the following conditions: When Q exceeds the preset normal fluctuation range When an anomaly is detected, the control system determines that there is an anomaly in the current locking mechanism and issues an alarm signal, while simultaneously recording the anomaly type and corresponding location information.

[0019] Furthermore, the control system is also equipped with a slipped tooth and float lock recognition module; the slipped tooth recognition module detects slipped teeth when the axial displacement continues to increase while the torque increase rate is below a first threshold. The condition is determined to be a stripped tooth fault, and the determination criteria are as follows: in The threshold value is the rate of change of axial displacement; the float lock identification module detects a rebound in axial displacement after the torque reaches a preset peak value, and the rebound amount exceeds the second threshold value. The fault is determined to be a float lock malfunction, and the determination criteria are as follows: and Where Speak is the axial displacement corresponding to the peak torque, and max(Spost) is the maximum value of the axial displacement after the locking is completed; the control system automatically performs a relocking operation or marks the workpiece as a defective product according to the identification result, and diverts it to the rework station through the conveyor line.

[0020] Furthermore, the CCD vision positioning unit includes at least two high-resolution industrial cameras and a coaxial light source. The industrial cameras are respectively positioned at the upstream station and directly above the fastening station of the fastening unit. The camera at the upstream station is used to acquire the global positioning information of the workpiece, and the camera directly above the fastening station is used to acquire the precise coordinates of the screw hole positions. The control system uses a sub-pixel edge extraction algorithm to perform center positioning of the screw hole positions, and the center coordinates of the hole positions are... Calculated using the weighted centroid method: in The coordinates of the edge points, The gradient magnitude weights are assigned to the corresponding edge points; and the hole coordinates are mapped to the motion coordinate system of the locking drive mechanism through coordinate transformation to achieve closed-loop position compensation of the locking unit. The positioning accuracy after compensation is better than ±0.05mm. The CCD vision positioning unit is independent of the screw posture detection camera set on the locking actuator. The screw posture detection camera is used to detect the axial sway angle of the self-tapping screw adsorbed on the bit assembly before locking, and controls the locking actuator to perform the screw throwing operation when the sway angle exceeds the preset threshold.

[0021] Furthermore, the production line includes multiple fastening units arranged in parallel along the conveyor line, with each unit independently performing its fastening task. The control system employs a dynamic task scheduling algorithm to dynamically allocate fastening tasks based on the real-time load status of each fastening unit, the screw hole distribution of the current workpiece, and the travel range of the fastening unit. The dynamic task scheduling algorithm is based on minimizing the total fastening time. Optimization objective: Where M is the number of locking units, and Nj is the number of screw holes allocated to the j-th locking unit. The estimated time for the jj-th fastening unit to complete the fastening of the i-th screw is calculated. By updating the task queue of each fastening unit in real time, multi-unit collaborative operation is achieved. When a fastening unit fails, the control system automatically reassigns its task to adjacent fastening units to ensure the continuous operation of the production line.

[0022] Furthermore, the control system also includes an adaptive parameter adjustment module. This module constructs a machine learning model based on historical fastening data. The machine learning model uses workpiece material, screw specifications, fastening depth, and ambient temperature as input feature vector X, and fastening speed as input feature vector X. The axial feed force Fz and torque threshold Tlimit are output parameters. During production, when the workpiece model is switched, the control system automatically calls the corresponding optimized parameter set to configure the locking unit. At the same time, the adaptive parameter adjustment module monitors the locking quality indicators in real time. When locking quality deviations occur at the same hole position of multiple consecutive workpieces, online parameter optimization is automatically triggered. The Bayesian optimization algorithm is used to find the optimal value within the preset parameter space. The optimization objective is to minimize the locking quality loss function L(p). Where p is the vector of parameters to be optimized. To measure the peak torque, For the target torque, For the final displacement deviation, This is a function to indicate if the lock failed to be locked. The weighting coefficient is used until the locking quality returns to stability.

[0023] The technical effects and advantages of this invention are as follows: This invention effectively reduces screw misalignment and prevents stripping by setting a guiding preload mechanism to radially guide the self-tapping screw and preload the workpiece before fastening. A nonlinear variable stiffness axial floating mechanism provides low-stiffness buffering at the initial stage of fastening and high-stiffness support at the end stage, improving the mechanical adaptability of the fastening process and achieving precise feeding. Joint torque and displacement detection constructs fastening quality characteristic parameters and weighted fusion indicators, accurately identifying abnormal states such as stripping and floating locks, improving the accuracy of fastening quality judgment. High-precision hole positioning is achieved through a CCD vision positioning unit, and screw sway detection is performed before fastening using a screw posture detection camera; this dual visual protection ensures fastening accuracy.

[0024] This invention enables automatic cyclic processing of workpieces between upper and lower conveyor belts through a cyclic conveyor system, improving the continuous operation capability and space utilization of the production line; through an adaptive parameter adjustment module, based on machine learning models and Bayesian optimization algorithms, it realizes automatic optimization of process parameters to adapt to the flexible production needs of multiple varieties and variable batches. Attached Figure Description

[0025] The following sections will describe some specific embodiments of the invention in detail by way of example and not limitation, with reference to the accompanying drawings. The same reference numerals in the drawings denote the same or similar parts or portions. Those skilled in the art should understand that these drawings are not necessarily drawn to scale. In the drawings: Figure 1 This is a schematic diagram of the structure of the present invention.

[0026] Figure 2 This is a schematic diagram of the internal structure of the present invention.

[0027] Figure 3 For the present invention Figure 2 A partially enlarged structural diagram of C. Figure 4 This is a flowchart of the dynamic task scheduling process of the present invention.

[0028] Figure 5 This is a flowchart illustrating the workflow of the cyclic conveying system of the present invention.

[0029] In the diagram: 1. First mounting plate placement box; 101. Vertical rail; 102. Conveying mechanism; 2. Locking execution mechanism; 201. Feeding unit; 202. Feed hopper; 203. Support plate; 204. First transverse slide rail; 205. First longitudinal slide block; 206. First longitudinal slide rail; 3. Vision positioning unit; 301. Second transverse slide rail; 302. Second longitudinal slide block; 303. Second longitudinal slide rail; 304. Second transverse fixing plate; 305. Cleaning mechanism; 4. Second mounting plate placement box; 5. Conveying mechanism; 501. First conveyor belt; 502. Second conveyor belt; 6. Control panel; 7. Mounting plate; 8. First fixing plate; 9. Locking unit; 901. Floating pressure head; 902. Locking drive mechanism; 903. Floating sleeve; 904. Bit assembly; 905. Guide pre-compression mechanism; 906. Axial floating mechanism; 10. Frame; 11. Drive mechanism. Detailed Implementation

[0030] The technical solutions of the embodiments of the present invention will be clearly and completely described below with reference to the accompanying drawings. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those skilled in the art without creative effort are within the scope of protection of the present invention.

[0031] Figure 1 This is a schematic diagram of the structure of the present invention. Figure 2 This is a schematic diagram of the internal structure of the present invention. Figure 3 For the present invention Figure 2 A magnified schematic diagram of the structure of C. Figure 4 This is a flowchart of the dynamic task scheduling process of the present invention. Figure 5 This is a flowchart illustrating the workflow of the cyclic conveying system of the present invention.

[0032] The solution of this embodiment provides a multi-unit self-tapping screw automatic fastening production line, including a frame 10, a conveyor line set on the frame, a feeding unit 201, a CCD vision positioning unit 3, and at least one fastening unit 9.

[0033] In this embodiment, the conveyor line is a circulating conveyor system, specifically including a frame 10, a first mounting plate placement box 1, and a second mounting plate placement box 4. The first mounting plate placement box 1 and the second mounting plate placement box 4 are respectively located at both ends of the frame 10, and each is equipped with a conveying mechanism 102, which can move up and down along the vertical rail 101. The transmission mechanism 5 includes a first transmission belt 501 and a second transmission belt 502, which are arranged vertically at intervals, and their ends are respectively corresponding to the first mounting plate placement box 1 and the second mounting plate placement box 4. The conveying mechanism 102 can receive and transmit the mounting plate 7 onto the first transmission belt 501 and the second transmission belt 502 for cyclic processing. The first mounting plate placement box 1 and the second mounting plate placement box 4 are respectively equipped with a drive mechanism 11, which is connected to the conveying mechanism 102 for driving the conveying mechanism 102 to move up and down along the vertical rail 101, so as to realize the cyclic switching of the mounting plate 7 between the first transmission belt 501 and the second transmission belt 502.

[0034] Both the locking actuator 2 and the CCD vision positioning unit 3 are mounted on the frame 10 and located above the first conveyor belt 501. The CCD vision positioning unit 3 is located upstream of the locking actuator 2 and is used to first locate the mounting holes on the mounting plate 7 and record the mounting plate number. The positioning data is transmitted and temporarily stored in the locking actuator 2. When the mounting plate 7 with the corresponding number moves to the locking actuator 2, locking is performed.

[0035] The fastening unit 9 includes a fastening drive mechanism 902 and a fastening execution mechanism 2. The fastening execution mechanism 2 includes a guide preload mechanism 905, an axial floating mechanism 906, and a bit assembly 904. The guide preload mechanism 905 includes a guide sleeve with a tapered inlet and a floating pressure head 901 coaxially arranged therewith. The floating pressure head 901 can generate a slight floating in the axial and radial directions relative to the fastening drive mechanism, which is used to radially guide the self-tapping screw and preload the workpiece before fastening. In the guide preload mechanism, an elastic centering component is provided between the floating pressure head 901 and the guide sleeve. The elastic centering component includes at least three elastic bodies evenly distributed in the circumference. The elastic bodies provide both radial restoring force and axial preload. The front end of the floating pressure head is provided with a contoured clamping surface adapted to the workpiece surface. A pressure sensor array is embedded in the contoured clamping surface for real-time detection of the contact pressure distribution on the workpiece surface during the preload stage.

[0036] The axial floating mechanism 906 is disposed between the locking drive mechanism 902 and the bit assembly 904, and is used to buffer and adjust the axial feed force during the locking process. The axial floating mechanism includes a floating sleeve 903, an elastic damping element disposed in the floating sleeve 903, and an axial displacement sensor. The elastic damping element is a nonlinear variable stiffness elastic body, and its stiffness changes continuously with the amount of compression.

[0037] The bit assembly 904 includes a bit body, a quick-change connector, and a bit wear detection unit 9041, which is used to drive the self-tapping screw to rotate and feed axially.

[0038] The locking actuator 2 specifically includes a feeding unit 201 and a feeding hopper 202. The feeding unit 201 is connected to the feeding hopper 202. A support plate 203, a first transverse slide rail 204, a first longitudinal slide block 205, and a first longitudinal slide rail 206 are arranged below the feeding hopper 202. The locking actuator 2 is connected to a first fixed plate 8 via the support plate 203. The first fixed plate 8 is fixedly connected to the first longitudinal slide block 205. The first longitudinal slide block 205 is slidably arranged with the first longitudinal slide rail 206 and is also movable along the length direction of the support plate 203 via the first transverse slide rail 204.

[0039] The CCD vision positioning unit 3 controls movement through the same structure as the locking actuator 2. Specifically, it is fixed by a second transverse slide rail 301, a second longitudinal slide block 302, a second longitudinal slide rail 303, and a second transverse fixing plate 304 to achieve independent movement control. The CCD vision positioning unit 3 also includes a cleaning mechanism 305, which is disposed on the second transverse fixing plate 304 and is used to clean the mounting holes on the mounting plate 7 before vision positioning.

[0040] The frame 10 is also equipped with a control panel 6, which is electrically connected to the control system. The control panel 6 displays the production line operating status, fastening quality data, and equipment fault information, and accepts parameter configuration input from the operator. The production line also includes a control system, which controls the movement path of the fastening unit based on the workpiece position information obtained by the CCD vision positioning unit, and simultaneously collects torque data and axial displacement data during the fastening process to determine the fastening status of the self-tapping screws. The control system is configured to construct a fastening quality characteristic parameter K based on the ratio of the torque change rate to the axial displacement change rate. Where T is the real-time torque, S is the axial displacement, and t is time; and through a preset threshold range Real-time classification and determination of the lock status.

[0041] During the pre-compression stage, the control system adjusts the pre-compression feed of the locking drive mechanism based on the feedback signal from the pressure sensor array, ensuring that the uniformity of contact pressure between the conformal clamping surface and the workpiece surface reaches a preset threshold. The contact pressure uniformity Up is defined as: in The standard deviation of the pressure values ​​collected by the pressure sensor array. The average pressure, when Pre-compression is completed at that time.

[0042] During the locking process, the axial floating mechanism adjusts the axial feed force in real time; the stiffness function k(x) of the elastic damping element is expressed as: Where x is the compression amount, k0 is the initial stiffness, k1 is the stiffness coefficient, and n is the nonlinear exponent, where n>1; the stiffness curve is configured to provide low stiffness buffering in the initial stage of locking and high stiffness support in the final stage of locking; the axial displacement sensor detects the axial displacement of the bit assembly in real time and feeds it back to the control system; the control system dynamically adjusts the axial feed speed of the locking drive mechanism according to the deviation between the real-time axial displacement and the preset displacement curve, so that the axial displacement tracking error is minimized. minimize.

[0043] The control system performs joint torque and displacement detection and constructs a locking quality judgment model. The model evaluates the locking state by calculating a weighted fusion index of the differential values ​​of real-time torque and axial displacement relative to the angle. The weighted fusion index is defined as follows: Where T is the real-time torque, Let S be the screw rotation angle, and S be the axial displacement. and The preset weighting coefficients are used, and they satisfy the following conditions: When Q exceeds the preset normal fluctuation range When an anomaly is detected, the control system determines that there is an anomaly in the current locking mechanism and issues an alarm signal, while simultaneously recording the anomaly type and corresponding location information.

[0044] The control system is also equipped with a slipped tooth and float lock recognition module; the slipped tooth recognition module detects slipped teeth when the axial displacement continues to increase while the torque increase rate is below a first threshold. The condition is determined to be a stripped tooth fault, and the determination criteria are as follows: in The threshold value is the rate of change of axial displacement; the float lock identification module detects a rebound in axial displacement after the torque reaches a preset peak value, and the rebound amount exceeds the second threshold value. The fault is determined to be a float lock malfunction, and the determination criteria are as follows: and Where Speak is the axial displacement corresponding to the peak torque, and max(Spost) is the maximum value of the axial displacement after the locking is completed; the control system automatically performs a relocking operation or marks the workpiece as a defective product according to the identification result, and diverts it to the rework station through the conveyor line, while recording the fault type, occurrence time and corresponding locking unit number, forming a production process quality traceability data chain.

[0045] The CCD vision positioning unit includes at least two high-resolution industrial cameras and a coaxial light source. The industrial cameras are respectively positioned at the upstream station and directly above the fastening station of the fastening unit. The camera at the upstream station is used to acquire the global positioning information of the workpiece, and the camera directly above the fastening station is used to acquire the precise coordinates of the screw hole positions. The control system uses a sub-pixel edge extraction algorithm to perform center positioning of the screw hole positions, and the center coordinates of the hole positions are... Calculated using the weighted centroid method: in The coordinates of the edge points, The gradient magnitude weights are assigned to the corresponding edge points; and the hole coordinates are mapped to the motion coordinate system of the locking drive mechanism through coordinate transformation to achieve closed-loop position compensation of the locking unit. The positioning accuracy after compensation is better than ±0.05mm. The CCD vision positioning unit is independent of the screw posture detection camera set on the locking actuator. The screw posture detection camera is used to detect the axial sway angle of the self-tapping screw adsorbed on the bit assembly before locking, and controls the locking actuator to perform the screw throwing operation when the sway angle exceeds the preset threshold.

[0046] The production line comprises multiple fastening units arranged in parallel along the conveyor line, each independently performing its fastening task. The control system employs a dynamic task scheduling algorithm to dynamically allocate fastening tasks based on the real-time load status of each fastening unit, the screw hole distribution of the current workpiece, and the travel range of the fastening unit. This dynamic task scheduling algorithm is based on minimizing the total fastening time. Optimization objective: Where M is the number of locking units, and Nj is the number of screw holes allocated to the j-th locking unit. The estimated time for the jj-th fastening unit to complete the fastening of the i-th screw is calculated. By updating the task queue of each fastening unit in real time, multi-unit collaborative operation is achieved. When a fastening unit fails, the control system automatically reassigns its task to adjacent fastening units to ensure the continuous operation of the production line.

[0047] The bit wear detection unit determines the degree of bit wear by detecting the fit clearance between the bit and the self-tapping screw drive slot. Specifically, it collects the torque fluctuation characteristics during bit rotation after each tightening and defines the torque fluctuation amplitude. for: when When the bit has reached its service life limit, the quick-change connector is automatically triggered to perform a bit replacement. A preset wear threshold is set. The control system simultaneously records the cumulative number of times the bit has been used and its wear trend, and predicts the remaining service life (RUL) of the bit using a linear regression model. in This is the predicted wear trend value under the current number of uses n. It provides an estimate of the wear rate and issues a maintenance reminder in advance when the RUL reaches the warning value.

[0048] The control system also includes an adaptive parameter adjustment module, which constructs a machine learning model based on historical fastening data. The machine learning model uses workpiece material, screw specifications, fastening depth, and ambient temperature as input feature vector X, and fastening speed as the parameter adjustment module. The axial feed force Fz and torque threshold Tlimit are output parameters. During production, when the workpiece model is switched, the control system automatically calls the corresponding optimized parameter set to configure the locking unit. At the same time, the adaptive parameter adjustment module monitors the locking quality indicators in real time. When locking quality deviations occur at the same hole position of multiple consecutive workpieces, online parameter optimization is automatically triggered. The Bayesian optimization algorithm is used to find the optimal value within the preset parameter space. The optimization objective is to minimize the locking quality loss function L(p). Where p is the vector of parameters to be optimized. To measure the peak torque, For the target torque, For the final displacement deviation, This is a function to indicate if the lock failed to be locked. The weighting coefficient is used until the locking quality returns to stability.

[0049] The working process of the multi-unit self-tapping screw automatic fastening production line provided in this embodiment is as follows: First, the operator places the mounting plate 7 to be processed into the first mounting plate placement box 1. The drive mechanism 11 drives the conveying mechanism 102 to rise along the vertical rail 101, conveying the mounting plate 7 onto the first conveyor belt 501. The first conveyor belt 501 transports the mounting plate 7 to below the CCD vision positioning unit 3.

[0050] The CCD vision positioning unit 3 moves along the second transverse slide rail 301, the second longitudinal slide block 302, and the second longitudinal slide rail 303. The cleaning mechanism 305 cleans the mounting hole. Then, the high-resolution industrial camera acquires the image of the mounting hole. The control system uses a sub-pixel edge extraction algorithm to calculate the center coordinates of the hole and temporarily stores the positioning data and the mounting plate number. The first conveyor belt 501 continues to transport the mounting plate 7 to below the locking actuator 2; the locking actuator 2 moves to above the target hole position through the first transverse slide rail 204, the first longitudinal slide block 205, and the first longitudinal slide rail 206 according to the temporarily stored positioning data. The locking actuator 2 descends, and the guide pre-pressure mechanism first contacts the workpiece surface. The conformal pressing surface at the front end of the floating pressure head fits against the workpiece surface. The pressure sensor array detects the contact pressure distribution in real time, and the control system adjusts the pre-pressure feed until the contact pressure uniformity reaches the threshold. The bit assembly picks up the self-tapping screw, and the screw attitude detection camera detects the screw yaw angle. If the yaw exceeds the threshold, the screw is thrown and the screw is picked up again. Then, the bit assembly drives the screw to rotate and feed axially. The axial floating mechanism dynamically adjusts the stiffness according to the fastening process. The control system synchronously collects torque and axial displacement data, calculates the fastening quality characteristic parameter K and the weighted fusion index Q, and determines the fastening status in real time. If a stripped tooth or floating lock fault is detected, the control system will automatically perform a relocking operation or mark the workpiece as a defective product and divert it to the rework station via the conveyor line. After the locking is completed, the mounting plate 7 is conveyed to the second mounting plate placement box 4, and then switched to the second conveyor belt 502 via the conveyor mechanism 102 to return to the first mounting plate placement box 1, realizing cyclic processing; The remote monitoring platform collects the operating data of each locking unit in real time, builds a digital twin model for visualization, and makes trend predictions on the overall efficiency of the equipment. When the predicted value is lower than the threshold, a maintenance work order is automatically generated.

[0051] Finally, it should be noted that the above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art can still modify the technical solutions described in the foregoing embodiments or make equivalent substitutions for some of the technical features. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A multi-unit self-tapping screw automatic fastening production line, characterized in that, Includes a frame, conveyor line, feeding unit, CCD vision positioning unit and at least one locking unit; The fastening unit includes a fastening drive mechanism and a fastening execution mechanism. The fastening execution mechanism includes a guide preload mechanism, which is used to radially guide the self-tapping screw and preload and position the workpiece before fastening. An axial floating mechanism is disposed between the locking drive mechanism and the bit assembly.

2. The multi-unit self-tapping screw automatic fastening production line according to claim 1, characterized in that, The production line also includes a control system. This control system controls the movement path of the fastening unit based on the workpiece position information acquired by the CCD vision positioning unit, and simultaneously collects torque and axial displacement data during the fastening process to determine the fastening status of the self-tapping screws. The control system is configured to construct a fastening quality characteristic parameter K based on the ratio of the torque change rate to the axial displacement change rate. Where T is the real-time torque, S is the axial displacement, and t is time; and through a preset threshold range Real-time classification and determination of the lock status.

3. The automatic fastening production line for multi-unit self-tapping screws according to claim 1, characterized in that, The axial floating mechanism includes a floating sleeve, an elastic damping element disposed within the floating sleeve, and an axial displacement sensor; the elastic damping element is a nonlinear variable stiffness elastic body, whose stiffness changes continuously with the compression amount, and the stiffness function k(x) is expressed as: Where x is the compression amount, k0 is the initial stiffness, k1 is the stiffness coefficient, and n is the nonlinear exponent, where n>1; the stiffness curve is configured to provide low stiffness buffering in the initial stage of locking and high stiffness support in the final stage of locking; the axial displacement sensor detects the axial displacement of the bit assembly in real time and feeds it back to the control system; the control system dynamically adjusts the axial feed speed of the locking drive mechanism according to the deviation between the real-time axial displacement and the preset displacement curve, so that the axial displacement tracking error is minimized. minimize.

4. The automatic fastening production line for multi-unit self-tapping screws according to claim 1, characterized in that, The control system performs joint torque and displacement detection and constructs a locking quality judgment model. The model evaluates the locking state by calculating a weighted fusion index of the differential values ​​of real-time torque relative to angle and axial displacement relative to angle. The weighted fusion index is defined as follows: Where T is the real-time torque, Let S be the screw rotation angle, and S be the axial displacement. and The preset weighting coefficients are used, and they satisfy the following conditions: When Q exceeds the preset normal fluctuation range When an anomaly is detected, the control system determines that there is an anomaly in the current locking mechanism and issues an alarm signal, while simultaneously recording the anomaly type and corresponding location information.

5. The automatic fastening production line for multi-unit self-tapping screws according to claim 1, characterized in that, The control system is also equipped with a slipped tooth and float lock recognition module; the slipped tooth recognition module detects slipped teeth when the axial displacement continues to increase while the torque increase rate is below a first threshold. The condition is determined to be a stripped tooth fault, and the determination criteria are as follows: in The threshold value is the rate of change of axial displacement; the float lock identification module detects a rebound in axial displacement after the torque reaches a preset peak value, and the rebound amount exceeds the second threshold value. The fault is determined to be a float lock malfunction, and the determination criteria are as follows: and Where Speak is the axial displacement corresponding to the peak torque, and max(Spost) is the maximum value of the axial displacement after the locking is completed; the control system automatically performs a relocking operation or marks the workpiece as a defective product according to the identification result, and diverts it to the rework station through the conveyor line.

6. The automatic fastening production line for multi-unit self-tapping screws according to claim 1, characterized in that, The CCD vision positioning unit includes at least two high-resolution industrial cameras and a coaxial light source. The industrial cameras are respectively positioned at the upstream station and directly above the fastening station of the fastening unit. The camera at the upstream station is used to acquire the global positioning information of the workpiece, and the camera directly above the fastening station is used to acquire the precise coordinates of the screw hole positions. The control system uses a sub-pixel edge extraction algorithm to perform center positioning of the screw hole positions, and the center coordinates of the hole positions are... Calculated using the weighted centroid method: in The coordinates of the edge points, The gradient magnitude weights are assigned to the corresponding edge points; and the hole coordinates are mapped to the motion coordinate system of the locking drive mechanism through coordinate transformation to achieve closed-loop position compensation of the locking unit. The positioning accuracy after compensation is better than ±0.05mm. The CCD vision positioning unit is independent of the screw posture detection camera set on the locking actuator. The screw posture detection camera is used to detect the axial sway angle of the self-tapping screw adsorbed on the bit assembly before locking, and controls the locking actuator to perform the screw throwing operation when the sway angle exceeds the preset threshold.

7. The multi-unit self-tapping screw automatic fastening production line according to claim 1, characterized in that, The production line comprises multiple fastening units arranged in parallel along the conveyor line, each independently performing its fastening task. The control system employs a dynamic task scheduling algorithm to dynamically allocate fastening tasks based on the real-time load status of each fastening unit, the screw hole distribution of the current workpiece, and the travel range of the fastening unit. This dynamic task scheduling algorithm is based on minimizing the total fastening time. Optimization objective: Where M is the number of locking units, and Nj is the number of screw holes allocated to the j-th locking unit. The estimated time for the jj-th fastening unit to complete the fastening of the i-th screw is calculated. By updating the task queue of each fastening unit in real time, multi-unit collaborative operation is achieved. When a fastening unit fails, the control system automatically reassigns its task to adjacent fastening units to ensure the continuous operation of the production line.

8. The automatic fastening production line for multi-unit self-tapping screws according to claim 1, characterized in that, The control system also includes an adaptive parameter adjustment module, which constructs a machine learning model based on historical fastening data. The machine learning model uses workpiece material, screw specifications, fastening depth, and ambient temperature as input feature vector X, and fastening speed as the parameter adjustment module. The axial feed force Fz and torque threshold Tlimit are output parameters. During production, when the workpiece model is switched, the control system automatically calls the corresponding optimized parameter set to configure the locking unit. At the same time, the adaptive parameter adjustment module monitors the locking quality indicators in real time. When locking quality deviations occur at the same hole position of multiple consecutive workpieces, online parameter optimization is automatically triggered. The Bayesian optimization algorithm is used to find the optimal value within the preset parameter space. The optimization objective is to minimize the locking quality loss function L(p). Where p is the vector of parameters to be optimized. To measure the peak torque, For the target torque, For the final displacement deviation, This is a function to indicate if the lock failed to be locked. The weighting coefficient is used until the locking quality returns to stability.

Images