A control method for waste yarn of extruder based on visual and pneumatic joint control

Abstract

The application is suitable for the technical field of metal wire production, and provides an extruding machine joint-passing waste wire control method based on visual and pneumatic joint control, which comprises the following steps: collecting continuous image information of the surface of the metal wire at the outlet of the extruding machine in real time; pre-processing the collected image information to output the position and shape information of the joint; calculating the action trigger timing and execution parameters to generate corresponding control instructions; opening the compressed air supply of the corresponding pneumatic execution mechanism according to the control instructions; separating the unqualified wire section containing the joint from the qualified wire; driving the sorting mechanism to guide the cut-off unqualified wire section to the collection area, and the qualified wire continues to advance along the production line; through the cooperative control of visual detection and pneumatic execution, the length of the joint-passing unqualified wire section can be controlled in a relatively short range, the waste of raw materials is reduced, the use efficiency of compressed air is optimized, and the automation level of the metal wire production line and the product quality stability are improved.

Description

Technical Field

[0001] This invention belongs to the field of metal wire production technology, and in particular relates to a method for controlling waste wire at the joint of an extruder based on vision and pneumatic joint control. Background Technology

[0002] In the continuous production of metal wires such as enameled aluminum flat wire and copper flat wire, the segmented metal rods need to be welded together before being fed into an extrusion press for forming. During the extrusion process, the welded joints will produce irreparable defects such as uneven structure, deformation, and cracks, which must be removed in subsequent processes; otherwise, they will seriously affect the electrical and mechanical properties of the final product.

[0003] Currently, the industry's methods for handling waste wire from over-joints mainly have the following problems: The manual processing suffers from insufficient precision and efficiency: manual visual inspection of joints is subject to subjectivity and fatigue, resulting in a high rate of missed and false detections. Furthermore, the reaction time of manual operation fluctuates, making it impossible to precisely control the cutting timing, leading to longer lengths of waste wire removed and significant waste of raw materials. Traditional detection methods have poor adaptability: metal detectors can only detect the presence of metal, but cannot distinguish between normal wires and welded joints, nor can they determine the precise range of joints. Small-sized joints are prone to being missed, and metal deposits on the surface of the wires can easily cause false detections, affecting the continuity of production. Pneumatic systems suffer from significant energy waste: Traditional pneumatic cutting and sorting devices maintain a continuous supply of compressed air, remaining in standby mode even during normal production periods without joint processing, resulting in substantial ineffective energy consumption and increased production costs. Lack of precise linkage between detection and execution: In existing methods, the detection and execution stages are independent of each other. They usually use fixed delay to control the execution action, which cannot dynamically adjust the cutting timing according to the real-time changes in the extruder's running speed, resulting in large deviations in the cutting position and unstable control of waste wire length. Low level of automation throughout the process: Existing methods require manual intervention in the identification, cutting confirmation and collection of waste filaments, and cannot achieve full automation from joint detection to waste filament separation, which limits the overall operating efficiency of the production line.

[0004] Therefore, a method for controlling waste wire at the extruder joint based on vision and pneumatic control is needed to solve the above problems. Summary of the Invention

[0005] The purpose of this invention is to provide a method for controlling waste wire at the joint of an extruder based on vision and pneumatic control, in order to solve the problems mentioned in the background art.

[0006] To achieve the above objectives, the present invention provides the following technical solution: A method for controlling waste wire at the joint of an extruder based on vision and pneumatic joint control includes the following steps: S1. Real-time acquisition of continuous image information of the surface of the metal wire at the extruder outlet.

[0007] By acquiring real-time image information of the metal wire surface, surface condition data can be continuously obtained, providing basic data support for subsequent joint identification. The non-contact image acquisition method does not damage the wire surface or affect the normal operation of the production line.

[0008] S2. Preprocess the acquired image information, identify the joint area on the metal wire using the trained recognition model, and output the position and shape information of the joint.

[0009] Image preprocessing improves image quality, highlights the differences between connectors and normal wires, and enhances the accuracy of subsequent identification. A trained identification model can automatically identify connector areas without manual intervention, improving efficiency and consistency. The output location and shape information provides precise information for subsequent cutting operations.

[0010] S3. Receive the position and shape information of the joint and the operating parameters of the extruder, calculate the action triggering time and execution parameters, and generate corresponding control commands.

[0011] By combining connector information with the extruder's real-time operating parameters, the optimal trigger timing can be dynamically calculated, avoiding cutting position deviations caused by extruder speed fluctuations. The generated control commands can precisely control the actions of the pneumatic actuators, ensuring the accuracy and timeliness of cutting and sorting operations.

[0012] S4. Start the compressed air supply to the corresponding pneumatic actuator according to the control command.

[0013] By employing an on-demand compressed air supply method, the compressed air supply is activated only when an action is required, effectively reducing the ineffective energy consumption of the pneumatic system and lowering production costs. Simultaneously, by precisely controlling the timing of compressed air supply, it is possible to ensure that the pneumatic actuator receives sufficient power when needed, guaranteeing the smooth completion of the action.

[0014] S5. When the connector reaches the preset cutting position, drive the cutting mechanism to perform the cutting operation, separating the unqualified wire segment containing the connector from the qualified wire.

[0015] By performing a cutting operation when the connector reaches a preset position, the cutting position can be precisely controlled, completely removing the defective wire segment including the connector while minimizing waste of qualified wire. The pneumatically driven cutting mechanism offers fast response, meeting the requirements of high-speed production lines.

[0016] S6. The drive sorting mechanism guides the cut-off unqualified wire segments to the collection area, while qualified wires continue to move forward along the production line.

[0017] The sorting mechanism automatically separates defective wire segments from qualified wire segments without manual intervention, improving the automation level of production. Defective wire segments are guided to a dedicated collection area for subsequent recycling and processing, while preventing defective wires from mixing with qualified products, thus ensuring product quality.

[0018] S7. After the non-conforming wire section is processed, shut off the compressed air supply to the corresponding pneumatic actuator and restore normal monitoring status.

[0019] Promptly shutting off the compressed air supply after the operation is completed can further reduce the energy consumption of the pneumatic system. The system automatically resumes normal monitoring, enabling continuous inspection of subsequent wires and ensuring production continuity.

[0020] Furthermore, in step S1, at least one set of imaging units is used to acquire images from at least one observation direction of the metal wire, while providing uniform and stable illumination to the imaging area and isolating it from interference from ambient light.

[0021] Multi-directional image acquisition allows for the acquisition of information about different surfaces of the metal wire, reducing blind spots and improving the comprehensiveness of joint identification. Uniform and stable illumination reduces the impact of surface reflections and shadows on image quality, improving image clarity and consistency. Isolating the system from ambient light interference ensures stable image quality under different lighting conditions, enhancing the system's adaptability.

[0022] Furthermore, in step S2, the preprocessing includes denoising and contrast adjustment of the original image; segmenting the metal wire region from the preprocessed image and extracting the joint features; classifying and locating the joint using a machine learning-based recognition model, and outputting the bounding box coordinates, size, and confidence information of the joint.

[0023] Denoising removal eliminates noise interference in images, improving the signal-to-noise ratio. Contrast adjustment enhances the differences between different regions in the image, highlighting joint features. Segmenting the metal wire region removes background interference, improving the accuracy and efficiency of identification. Machine learning-based identification models can learn from a large amount of sample data, adapting to metal wires of different specifications and surface conditions, thus improving the accuracy and adaptability of joint identification. The detailed output provides precise information for subsequent control.

[0024] Furthermore, in step S3, the operating parameters include the real-time operating speed of the extruder and the production line status; the precise time for the joint to reach the cutting position is calculated based on the position of the joint in the image, the distance between the imaging position and the cutting position, and the operating speed of the extruder; and cutting control commands and sorting control commands are generated based on the calculation results.

[0025] By acquiring real-time operating parameters of the extruder, the operating status of the production line can be dynamically monitored, providing an accurate basis for calculating the timing of actions. Calculating the arrival time based on the joint position and operating speed enables precise control of the cutting timing, avoiding cutting position deviations caused by speed variations. Generating separate cutting and sorting control commands allows for independent control and precise coordination of the two actions, improving operational accuracy.

[0026] Furthermore, in step S4, the compressed air supply to the corresponding pneumatic actuator is turned on by controlling the on / off state of the solenoid valve group; the pressure status of the compressed air pipeline is monitored in real time, and an alarm signal is issued when the pressure is lower than the set threshold; the flow parameters of the compressed air are adjusted according to the working requirements of the pneumatic actuator.

[0027] The solenoid valve assembly allows for precise control of the compressed air supply to each pneumatic actuator, enabling on-demand air supply. Real-time pressure monitoring can promptly detect any abnormalities in the compressed air system, ensuring that the pneumatic actuators receive sufficient power. Adjusting flow parameters provides the appropriate compressed air flow rate according to the working requirements of different pneumatic actuators, ensuring smooth operation while avoiding unnecessary energy consumption.

[0028] Furthermore, in step S5, a cylinder is used to drive the cutting component to cooperate with the support to complete the cutting operation; the real-time position of the metal wire is detected by a position sensor, and the cutting action is triggered when the joint is detected to reach the cutting position; after the cutting is completed, the cutting component automatically resets.

[0029] The cylinder drive features fast response and stable driving force, enabling rapid completion of the cutting operation. The cooperation between the cutting element and the support base ensures a smooth cut of the wire, guaranteeing the quality of the cut surface. The position sensor provides real-time position feedback, forming a closed-loop control that improves the accuracy and reliability of the cutting action. Automatic resetting of the cutting element prepares it for the next cutting operation, ensuring continuous production.

[0030] Furthermore, in step S6, a cylinder is used to drive the guide to swing, changing the direction of movement of the unqualified wire segments and guiding them to the collection tank; qualified wires continue to move forward along the original path; after sorting is completed, the guide is automatically reset.

[0031] The cylinder-driven guide oscillation function can quickly change the direction of wire movement, achieving rapid separation of defective and qualified wires. The collection trough can collect defective wire segments for easy subsequent recycling. Automatic guide reset ensures the normal transport of qualified wires, without affecting the continuous operation of the production line.

[0032] Furthermore, in step S7, after both the cutting mechanism and the sorting mechanism have completed their reset actions, the compressed air supply is turned off after a preset time; relevant data for this waste wire processing is recorded, including the joint position, waste wire length, and processing time.

[0033] Delaying the shutdown of compressed air supply ensures the pneumatic actuator fully resets, preventing incomplete operation due to premature shutdown. Recording relevant data provides support for production management and process improvement, enabling companies to understand production status and optimize processes.

[0034] Furthermore, it also includes the following steps: S8. Human-machine interaction: Real-time display of system operating status, joint detection records and waste wire statistics; receives system parameter setting instructions and manual intervention instructions input by operators; issues audible and visual alarms and displays abnormal information when system abnormalities occur.

[0035] Real-time display of system operating information allows operators to keep abreast of production conditions, facilitating production management. Parameter setting functions allow operators to adjust system parameters according to different production needs, improving system adaptability. Manual intervention functionality enables operators to take control in special circumstances, ensuring safe and stable production. Audible and visual alarm functions promptly alert operators to abnormal situations, reducing production losses caused by malfunctions.

[0036] Furthermore, it also includes the following steps: S9. Anomaly Handling: When the confidence level of the connector output by the recognition model is lower than the set threshold, adjust the image acquisition parameters and perform secondary recognition; when the expected connector is not detected, issue a missed detection alarm and prompt manual confirmation; when the pneumatic actuator operates abnormally, issue a fault alarm and take corresponding safety protection measures.

[0037] Secondary identification improves the accuracy of connector identification and reduces false positives and false negatives. False negative alarms promptly alert operators to check for missed connectors, preventing defective products from entering the next process. Fault alarms and safety protection measures enable timely action in case of system malfunctions, protecting personnel and equipment and minimizing losses.

[0038] Compared with the prior art, the beneficial effects of the present invention are: This invention improves the accuracy and adaptability of connector identification: by using image acquisition technology and machine learning recognition algorithms, it can obtain detailed feature information of the surface of metal wires. By learning from a large number of labeled samples, it can adapt to metal wires of different specifications and surface conditions, effectively reducing the occurrence of missed detections and false detections, and improving the reliability of connector identification. This invention precisely controls the length of waste wire, reducing raw material waste: by collecting extruder operating parameters in real time and dynamically calculating the timing of actions, combined with the rapid response of the pneumatic actuator, the length of the defective wire segment after splicing can be controlled within a shorter range. Compared with traditional processing methods, this can reduce a significant amount of raw material waste and improve the economic benefits of enterprises. This invention optimizes the efficiency of compressed air use and reduces production energy consumption: by adopting a control method of supplying compressed air on demand, the compressed air is turned on only when cut-off and sorting actions are required, and kept off at other times, which can effectively reduce the ineffective energy consumption of the pneumatic system and save production costs. This invention achieves precise joint control of vision and pneumatics, improving system stability: It organically combines visual detection and pneumatic execution, and achieves precise coordination between the two through real-time timing calculation. It can automatically adjust the timing of actions according to the changes in the running speed of the extruder, ensuring the accuracy of the cutting position. At the same time, the position feedback and abnormal handling mechanism further improves the stability and reliability of the system operation. This invention improves the level of production automation and reduces labor intensity: it realizes full-process automation from joint detection and cutting to sorting, eliminating the need for manual intervention in the identification and handling of waste wires, reducing the labor intensity of workers, improving the overall operating efficiency of the production line, and supporting long-term continuous production.

[0039] To more clearly illustrate the structural features and effects of the present invention, the present invention will be described in detail below with reference to the accompanying drawings and specific embodiments. Attached Figure Description

[0040] Figure 1 This is a schematic diagram of the overall process of the present invention. Detailed Implementation

[0041] To make the objectives, technical solutions, and advantages of this invention clearer, the invention will be further described in detail below with reference to the accompanying drawings and embodiments. It should be understood that the specific embodiments described herein are merely illustrative and not intended to limit the invention.

[0042] The specific implementation of the present invention will be described in detail below with reference to specific embodiments.

[0043] Example 1 like Figure 1As shown, this embodiment of the invention provides a method for controlling waste wire at the joint of an extruder based on vision and pneumatic joint control. It is applicable to continuous production scenarios of conventional specification metal wires. The specific steps are as follows: S1. Two imaging units are set at the extruder outlet, forming orthogonal observation angles from above and to the side of the metal wire, respectively, to continuously acquire image information of the metal wire surface at a preset frame rate. At the same time, an illumination unit is set around the lens of each imaging unit to provide uniform and soft illumination for the imaging area; a light-shielding unit is set outside the imaging unit and the illumination unit to isolate interference from external ambient light such as workshop lights and sunlight, ensuring that the acquired images are clear and stable.

[0044] S2. Preprocessing of the acquired raw images: First, a filtering algorithm is used to remove noise interference from the image. Then, a contrast enhancement algorithm is used to enhance the image contrast, highlighting the difference between the joint area and the normal wire area. Next, an image segmentation algorithm is used to segment the metal wire area from the background and remove background interference. Finally, texture, contour, and other feature information of the wire area are extracted and input into a pre-trained machine learning-based recognition model. The recognition model analyzes the input features, identifies the position and shape of the welded joint, and outputs the bounding box coordinates, size, and confidence information of the joint. When a joint with a confidence level higher than a set threshold is identified, the relevant information is transmitted to the control unit.

[0045] S3. The control unit receives the joint position and shape information from the identification module, and simultaneously obtains parameters such as real-time operating speed and production line status from the extruder control system. Based on the joint's longitudinal coordinates in the image, the fixed distance between the imaging unit and the cutting mechanism, and the extruder's real-time operating speed, it calculates the precise time for the joint to reach the cutting position. Based on the calculation results, it generates cutting control commands and sorting control commands in advance and determines the command transmission time.

[0046] S4. Before the connector reaches the cut-off position, the control unit sends a command to the air source management unit to open the solenoid valves corresponding to the cut-off cylinder and the sorting cylinder, providing compressed air to the pneumatic actuator. The air source management unit monitors the pressure status of the compressed air main pipeline in real time, and issues an audible and visual alarm when the pressure is lower than the set threshold; at the same time, it adjusts the flow rate of compressed air through the flow regulating valve according to the working requirements of the pneumatic actuator to ensure that the pneumatic actuator can operate stably.

[0047] S5. When the position sensor detects that the front end of the metal wire has reached the cutting position, and the calculated arrival time of the joint matches the actual detection time, the control unit sends an action command to the cutting cylinder. The cutting cylinder extends, driving the cutting element to cooperate with the support seat, cutting the defective wire segment containing the joint from the qualified wire. After cutting is completed, the cutting cylinder immediately retracts automatically, driving the cutting element to reset.

[0048] S6. After a preset time following the completion of the cutting action, the control unit sends an action command to the sorting cylinder. The sorting cylinder extends, driving the guide to swing downwards, changing the direction of movement of the defective wire segments and causing them to fall into the collection trough below. The qualified wire segments continue to be conveyed forward along the original path. Once the defective wire segments have completely fallen into the collection trough, the sorting cylinder automatically retracts, causing the guide to reset.

[0049] S7. After both the cutting-off cylinder and the sorting cylinder have completed their reset actions, after a preset delay, the control unit sends a command to the air source management unit to close the solenoid valves corresponding to the cutting-off cylinder and the sorting cylinder, cutting off the compressed air supply. Simultaneously, the control unit records relevant data for this waste wire processing, including the joint detection time, location, waste wire length, processing time, etc., and stores the data in the database. The system returns to normal monitoring status, continues to acquire images, and detects the next joint.

[0050] This embodiment effectively controls the length of waste wire at the joint by acquiring orthogonal viewpoint images, recognizing through machine learning, and controlling precise timing, while optimizing the efficiency of compressed air usage, thereby improving production efficiency and economic benefits.

[0051] Example 2 The difference between this embodiment and Embodiment 1 is that it is applicable to metal wire production scenarios with small specifications and high precision requirements. The specific adjustments are as follows: In step S1, three imaging units are used to acquire images from above and two sides of the metal wire, forming a comprehensive observation system to further reduce blind spots and improve the recognition accuracy of small-sized joints and edge joints. Simultaneously, a high-resolution imaging unit and a high-brightness illumination unit are employed to enhance the detail representation of the images.

[0052] In step S2, an improved target detection algorithm is used for joint identification. This algorithm adds a multi-scale feature fusion module and an attention mechanism, which can better detect small-sized and blurry targets. At the same time, an appropriate recognition confidence threshold is set to reduce the occurrence of false detections.

[0053] In step S3, compensation calculations for linear speed fluctuations are added. When the extruder's operating speed changes, the time it takes for the joint to reach the cutting position is updated in real time to ensure the accuracy of the cutting position. At the same time, the pre-cutting length is dynamically adjusted based on the joint's size information to minimize the length of waste wire while ensuring complete removal of the joint.

[0054] This embodiment can effectively identify minute welding joints, improve the accuracy of waste wire length control, and meet the production requirements of high-precision products.

[0055] Example 3 The difference between this embodiment and Embodiment 1 is that it is applicable to the production scenarios of large-diameter, high-hardness metal wires. The specific adjustments are as follows: In step S4, an air storage unit is installed on the air supply branch of the cutting cylinder to store a certain amount of compressed air, ensuring sufficient power is provided when cutting large-diameter wires and avoiding incomplete cutting due to insufficient compressed air pressure. Simultaneously, the working pressure of the compressed air is adjusted according to the wire specifications to enhance the cutting force.

[0056] In step S5, a dual-cylinder driven cutting mechanism is adopted, with the two cylinders operating synchronously to provide greater cutting force. Simultaneously, a buffer mechanism is installed between the cutting workpiece and the cylinders to absorb the impact force during the cutting process, reducing wear between the cutting workpiece and the support base, and extending service life.

[0057] In step S7, the wear condition of the cutting component is monitored. Based on the cumulative number of cuts and the quality of waste wire, the operator is automatically prompted to replace the cutting component to avoid a decrease in cutting quality due to wear of the cutting component.

[0058] This embodiment can stably cut high-hardness metal wires with large cross-sectional areas, ensuring the quality of the cut surface and extending the service life of the equipment.

[0059] Example 4 The difference between this embodiment and Embodiment 1 is that data statistics and analysis functions have been added. The specific steps are as follows: Following step S7, a data statistics and analysis step is added: The stored waste wire processing data is periodically statistically analyzed to generate reports for different time dimensions, including statistics on splice frequency, average waste wire length, equipment operating efficiency, energy consumption, and other indicators. Through data analysis, abnormal situations in the production process are identified, such as an abnormally high splice frequency during a certain period, prompting process personnel to make adjustments in a timely manner.

[0060] Simultaneously, a joint defect database is established to classify and statistically analyze different types of joint defects, identifying their causes and providing data support for optimizing welding and extrusion processes. Through continuous data analysis and process improvement, the generation of joints can be further reduced, lowering the waste wire rate.

[0061] This embodiment can help enterprises achieve refined production management and continuously improve production efficiency and product quality.

[0062] Example 5 The difference between this embodiment and Embodiment 1 is that remote monitoring and fault diagnosis functions have been added. The specific steps are as follows: The system transmits operational data to a cloud server in real time, allowing managers to remotely monitor the production line's status, view real-time production data, joint inspection records, and alarm information via terminal devices. When an anomaly occurs, the system automatically sends an alarm to the manager's terminal device, facilitating timely troubleshooting.

[0063] Simultaneously, a fault diagnosis model is established to analyze fault data generated during system operation, identify fault types and causes, and provide corresponding solutions. Technicians can then remotely diagnose and maintain the system, reducing on-site maintenance time and costs.

[0064] This embodiment is applicable to large, decentralized production enterprises, enabling centralized monitoring and management of multiple production bases.

[0065] Example 6 The difference between this embodiment and Embodiment 1 is that a waste filament compaction step is added, as detailed below: Following step S6, a waste wire compaction step is added: when the waste wire in the collection tank accumulates to a certain amount, the compaction cylinder is driven to move the compaction plate downwards, compacting the loose waste wire. After compaction is completed, the compaction plate automatically resets. When the waste wire is compacted to a set volume, a prompt signal is issued to remind the operator to clean the collection tank in a timely manner.

[0066] The volume of compacted waste filaments can be significantly reduced, lowering transportation and storage costs and facilitating subsequent recycling.

[0067] This embodiment is suitable for production enterprises that generate a large amount of waste filaments, and can improve the efficiency of waste filament treatment and reduce the cost of waste filament treatment.

[0068] Example 7 The difference between this embodiment and Embodiment 1 is that a multi-type wire adaptive function has been added. The specific steps are as follows: The system pre-sets production parameter templates for metal wires of different specifications and materials, including image acquisition parameters, recognition model parameters, cutting parameters, and pneumatic parameters. When changing product types, operators only need to select the corresponding parameter template, and the system will automatically adjust all relevant parameters without requiring manual settings.

[0069] Meanwhile, the system has a self-learning function, continuously optimizing the recognition model and control parameters during the production process to improve its adaptability to new types of wires. When a new connector type appears during production, the system can automatically add it to the training samples, update the recognition model, and improve the recognition accuracy.

[0070] This embodiment can quickly adapt to a multi-variety, small-batch production mode, reduce changeover time, and improve production efficiency.

[0071] Example 8 The difference between this embodiment and Embodiment 1 is that a safety interlock protection function has been added. The specific steps are as follows: A safety interlock mechanism is installed in the system. When an operator opens the equipment's protective door or enters a hazardous area, the system immediately stops all pneumatic actuators and cuts off the compressed air supply to prevent accidents. Simultaneously, manual operation buttons are installed on the pneumatic actuators for easy manual operation during equipment commissioning and maintenance.

[0072] When the system detects an emergency stop signal, it immediately stops all operations and cuts off the compressed air supply and the operation of the extruder to ensure the safety of personnel and equipment.

[0073] This embodiment can effectively improve system security and prevent security incidents from occurring.

[0074] The above description is only a preferred embodiment of the present invention and is not intended to limit the present invention. Any modifications, equivalent substitutions, and improvements made within the spirit and principles of the present invention should be included within the protection scope of the present invention.

Claims

1. A method for controlling waste wire at the joint of an extruder based on vision and pneumatic joint control, characterized in that, Includes the following steps: S1. Real-time acquisition of continuous image information of the surface of the metal wire at the extruder outlet; S2. Preprocess the acquired image information, identify the joint area on the metal wire using the trained recognition model, and output the position and shape information of the joint. S3. Receive the position and shape information of the joint and the operating parameters of the extruder, calculate the action triggering time and execution parameters, and generate corresponding control commands; S4. Start the compressed air supply to the corresponding pneumatic actuator according to the control command; S5. When the connector reaches the preset cutting position, drive the cutting mechanism to perform the cutting operation to separate the unqualified wire segment containing the connector from the qualified wire. S6. The drive sorting mechanism guides the cut-off unqualified wire segments to the collection area, while qualified wires continue to move along the production line. S7. After the non-conforming wire section is processed, shut off the compressed air supply to the corresponding pneumatic actuator and restore normal monitoring status.

2. The method for controlling waste wire at the joint of an extruder based on vision and pneumatic joint control according to claim 1, characterized in that, In step S1, at least one set of imaging units is used to acquire images from at least one observation direction of the metal wire, while providing uniform and stable illumination to the imaging area and isolating it from interference from ambient light.

3. The method for controlling waste wire at the joint of an extruder based on vision and pneumatic joint control according to claim 1, characterized in that, In step S2, the preprocessing includes denoising and contrast adjustment of the original image; segmenting the metal wire region from the preprocessed image and extracting the joint features; classifying and locating the joint using a machine learning-based recognition model, and outputting the bounding box coordinates, size, and confidence information of the joint.

4. The method for controlling waste wire at the joint of an extruder based on vision and pneumatic joint control according to claim 1, characterized in that, In step S3, the operating parameters include the real-time operating speed of the extruder and the production line status; the precise time for the joint to reach the cutting position is calculated based on the position of the joint in the image, the distance between the imaging position and the cutting position, and the operating speed of the extruder; and cutting control commands and sorting control commands are generated based on the calculation results.

5. The method for controlling waste wire at the joint of an extruder based on vision and pneumatic joint control according to claim 1, characterized in that, In step S4, the compressed air supply to the corresponding pneumatic actuator is turned on by controlling the on / off state of the solenoid valve group; the pressure status of the compressed air pipeline is monitored in real time, and an alarm signal is issued when the pressure is lower than the set threshold; the flow parameters of compressed air are adjusted according to the working requirements of the pneumatic actuator.

6. The method for controlling waste wire at the joint of an extruder based on vision and pneumatic joint control according to claim 1, characterized in that, In step S5, a cylinder drives the cutting component to cooperate with the support to complete the cutting operation; a position sensor detects the real-time position of the metal wire, and the cutting action is triggered when the joint is detected to reach the cutting position. The cut piece automatically resets after the cutting is completed.

7. The method for controlling waste wire at the joint of an extruder based on vision and pneumatic joint control according to claim 1, characterized in that, In step S6, a cylinder is used to drive the guide to swing, changing the direction of movement of the unqualified wire segments and guiding them to the collection tank; qualified wires continue to move forward along the original path; after sorting is completed, the guide is automatically reset.

8. The method for controlling waste wire at the joint of an extruder based on vision and pneumatic joint control according to claim 1, characterized in that, In step S7, after both the cutting mechanism and the sorting mechanism have completed their reset actions, the compressed air supply is turned off after a preset time; relevant data for this waste wire processing is recorded, including the joint position, waste wire length, and processing time.

9. The method for controlling waste wire at the joint of an extruder based on vision and pneumatic joint control according to claim 1, characterized in that, It also includes the following steps: S8. Human-machine interaction: Real-time display of system operating status, joint detection records and waste wire statistics; receives system parameter setting instructions and manual intervention instructions input by operators; issues audible and visual alarms and displays abnormal information when system abnormalities occur.

10. The method for controlling waste wire at the joint of an extruder based on vision and pneumatic joint control according to claim 1, characterized in that, It also includes the following steps: S9. Anomaly Handling: When the confidence level of the connector output by the recognition model is lower than the set threshold, adjust the image acquisition parameters and perform secondary recognition; when the expected connector is not detected, issue a missed detection alarm and prompt manual confirmation; when the pneumatic actuator operates abnormally, issue a fault alarm and take corresponding safety protection measures.

Images