A sewing machine thread breakage detection, control and emergency response system and its computer equipment

Abstract

This invention relates to the field of sewing machine technology, specifically to a sewing machine thread breakage detection, control, and emergency response system and its computer equipment. The system includes a thread breakage detection module comprising a photoelectric sensor, a tension sensor, a Hall effect speed sensor, and a visual recognition unit. It is used to collect and comprehensively identify the presence, tension changes, motion continuity, and morphology of the sewing thread in real time. This invention employs a multi-source detection architecture that deeply integrates photoelectric sensors, tension sensors, Hall effect speed sensors, and a visual recognition unit. Through contour detection algorithms, deep learning semantic segmentation technology, and a graded response mechanism for thread breakage and micro-damage, coupled with a multi-interface data storage and networking module, it solves problems such as high false alarm rates and delayed response in sewing machine thread breakage detection through multi-dimensional signal cross-validation, accurate identification of latent faults, differentiated emergency response, and full lifecycle data traceability. This improves the accuracy and reliability of thread breakage and fault detection in sewing machines.

Description

Technical Field

[0001] This invention relates to the field of sewing machine technology, specifically to a sewing machine thread breakage detection, control and emergency response system and its computer equipment. Background Technology

[0002] As the core equipment in the production of textiles such as clothing, footwear, and bags, sewing machines directly affect product costs and production efficiency through their working efficiency and sewing quality. Sewing machines often operate in a long-term continuous operation mode, and the sewing thread is easily affected by factors such as friction, stretching, and wear, which can lead to thread breakage or hidden micro-damage.

[0003] Some sewing machine thread breakage detection methods rely on traditional mechanical thread-stopping devices, which trigger a mechanical switch based on thread tension to achieve detection. This method has a slow response time, a high false alarm rate, and cannot identify micro-damage to the thread, often causing the sewing machine to continue running after a thread breakage, resulting in fabric waste and rework. Other methods use a single sensor detection solution, such as using only a photoelectric sensor or a tension sensor. Although this improves the timeliness of detection to some extent, it is affected by factors such as thread vibration, ambient light interference, and fabric reflection, resulting in insufficient detection accuracy and a lack of early warning capabilities for hidden faults such as thread fatigue and wear. Summary of the Invention

[0004] The purpose of this invention is to solve the problems of high false alarm rate, delayed response, and damage warning in the existing technology for sewing machine thread breakage detection.

[0005] To achieve the above objectives, the present invention provides a sewing machine thread breakage detection, control, and emergency response system, the system comprising:

[0006] The thread breakage detection module includes a photoelectric sensor, a tension sensor, a Hall speed sensor, and a vision recognition unit, which is used to collect and comprehensively identify the presence, tension changes, motion continuity, and morphology of the sewing thread in real time.

[0007] The control and action execution module has its input terminal electrically connected to the output terminal of the thread breakage detection module. It is used to receive the thread breakage judgment signal and output the stop control command, braking command or alarm command to realize the automatic control and emergency execution of the sewing machine's operating status.

[0008] The human-computer interaction module is used to display and prompt the thread breakage status, fault type, and sewing machine operating status in real time, and to receive manual operation input;

[0009] The data storage and networking module communicates bidirectionally with the thread breakage detection module and the control and action execution module, respectively, to record and store the thread breakage status, sensor data, operation records and sewing machine operating status, and uploads the data to external devices or network systems through the communication interface to realize remote monitoring and data management;

[0010] The thread breakage detection module fuses the signals from each sensor with the visual recognition results for judgment. When the preset thread breakage conditions are met, it outputs a thread breakage judgment signal. The preset thread breakage conditions include photoelectric signal abnormality lasting for more than or equal to 10ms, tension signal exceeding 0.3N-3N, wheel speed below 50r / min, and visual recognition of thread interruption length greater than or equal to 2mm.

[0011] Furthermore, the photoelectric sensor is positioned above the sewing thread transmission path to detect whether the sewing thread is blocking light and output a presence signal;

[0012] The tension sensor is connected in series in the suture transmission mechanism to detect the working tension of the suture and output a tension change signal;

[0013] The Hall speed sensor is installed on the end of the wheel shaft on the suture transmission path to collect the wheel speed for characterization. When the speed is lower than the third preset threshold, it is determined that the suture movement is interrupted.

[0014] Furthermore, in the thread breakage detection module, the visual recognition unit identifies the thread trajectory, slack state, and breakage state based on the image captured by the camera.

[0015] The disconnection detection module performs fusion judgment on the above signals, and outputs the disconnection judgment result when any signal meets the abnormal judgment condition or the recognition result is a disconnection state.

[0016] Furthermore, the control and action execution module is electrically connected to the disconnection detection module. The control and action execution module includes a controller, a motor drive unit, a braking execution unit, and an alarm output unit. The controller is electrically connected to the motor drive unit, the braking execution unit, and the alarm output unit, respectively.

[0017] Furthermore, in the control and action execution module, the controller is used to perform logical processing and state decision on the thread breakage judgment signal; the motor drive unit is used to control the sewing machine main motor to stop and decelerate; the braking execution unit is used to brake or lock the sewing machine needle bar mechanism when a thread breakage occurs; the alarm output unit is used to execute audible and visual alarms or status prompts; and the human-machine interaction module includes a display unit and an input unit. The display unit is used to display the thread breakage location, thread breakage type, and fault status information; and the input unit is used to receive manual input to perform reset, power on / off, and mode switching operations.

[0018] Furthermore, the visual recognition unit is also used for suture breakage identification and suture micro-damage grading, and the visual recognition unit includes:

[0019] The image acquisition subunit is used to continuously acquire images of the sewing thread's trajectory using a built-in high-definition industrial camera.

[0020] The supplementary light subunit is used to eliminate ambient light interference and obtain clear dynamic images of the sutures;

[0021] The image preprocessing subunit is used to perform grayscale conversion, Gaussian filtering for noise reduction, and edge enhancement on the acquired image.

[0022] The image segmentation subunit is used to separate the suture from the background and extract the effective pixels of the suture region using a threshold segmentation algorithm.

[0023] The contour detection subunit is used to identify the continuous shape of the suture. When the suture contour is interrupted and the interruption length is ≥2mm, it outputs a suture break recognition signal.

[0024] The deep learning processing subunit is used to perform pixel-level segmentation of the suture region using a semantic segmentation algorithm, extract feature parameters, and identify the micro-damage state based on the feature parameters and classify it into first-level micro-damage, second-level micro-damage, or third-level micro-damage. The feature parameters include fiber breakage density, surface roughness, and wire diameter change rate.

[0025] Furthermore, the data storage and networking module includes a storage unit and a communication unit. The storage unit is used to store disconnection detection data, fault records, and historical operating parameters. The communication unit includes a wired communication interface and a wireless communication interface.

[0026] Furthermore, the storage unit is used to store the image data, micro-damage grading data and processing results of each step output by the visual recognition unit, and is classified and stored according to the format of timestamp and detection type; the storage unit is also used to pre-store the training model parameters of the deep learning semantic segmentation algorithm, support online model updates and local self-calibration, and receive update packets through the communication unit and automatically complete calibration;

[0027] The communication unit includes an RS485 wired communication interface, a WiFi wireless communication interface, and a 4G / 5G communication module, which supports uploading suture micro-damage data, suture breakage detection data, statistical analysis reports, and micro-damage frequency to the remote management system.

[0028] Furthermore, the visual recognition unit is also used for early warning and control based on the suture micro-damage grading, and the visual recognition unit includes:

[0029] The first-level micro-damage control subunit is used to send a prompt command to the human-machine interaction module of the motor drive unit when the suture micro-damage is classified as first-level micro-damage, without changing the operating parameters and maintaining the normal working state.

[0030] The secondary micro-damage control subunit is used to output a speed control command to the motor drive unit when the suture micro-damage is classified as secondary micro-damage, reducing the sewing machine speed to half of the original rated speed; and to output intermittent audible and visual prompts to the alarm output unit to trigger periodic alarm prompts.

[0031] The three-level micro-damage control subunit is used to output a stop control command to the motor drive unit and a braking command to the braking execution unit when the suture micro-damage is classified as level two micro-damage.

[0032] On the other hand, it includes a processor, a memory, and a computer program stored in the memory, wherein the processor executes the computer program and applies it to the above-described sewing machine thread breakage detection control and emergency response system.

[0033] Compared with known public technologies, the technical solution provided by this invention has the following beneficial effects:

[0034] This invention employs a multi-source detection architecture that deeply integrates photoelectric sensors, tension sensors, Hall effect speed sensors, and visual recognition units. Through contour detection algorithms, deep learning semantic segmentation technology, and a graded response mechanism for thread breakage and micro-damage, coupled with a multi-interface data storage and networking module, it solves the problems of high false alarm rate, delayed response, and damage warning in sewing machine thread breakage detection by multi-dimensional signal cross-verification, accurate identification of hidden faults, differentiated emergency handling, and full lifecycle data traceability. This improves the accuracy and reliability of thread breakage and fault detection in sewing machines. Attached Figure Description

[0035] Figure 1 This is an overall system diagram of a sewing machine thread breakage detection, control, and emergency response system according to the present invention;

[0036] Figure 2 This is a system operation logic flowchart of a sewing machine thread breakage detection, control and emergency response system according to the present invention;

[0037] Figure 3 This is a system diagram of the visual recognition unit of a sewing machine thread breakage detection, control and emergency response system according to the present invention. Detailed Implementation

[0038] To enable those skilled in the art to better understand the present invention, the technical solutions of the present invention will be clearly and completely described below with reference to the accompanying drawings of the embodiments of the present invention. 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 should fall within the scope of protection of the present invention.

[0039] It should be noted that the terms "first," "second," etc., in the specification, claims, and accompanying drawings of this invention are used to distinguish similar objects and are not necessarily used to describe a specific order or sequence. It should be understood that such data can be interchanged where appropriate so that the embodiments of the invention described herein can be implemented in orders other than those illustrated or described herein. Furthermore, the terms "comprising" and "having," and any variations thereof, are intended to cover a non-exclusive inclusion; for example, a process, method, system, product, or apparatus that comprises a series of steps or units is not necessarily limited to those steps or units explicitly listed, but may include other steps or units not explicitly listed or inherent to such processes, methods, products, or apparatus.

[0040] The present invention will now be described in further detail with reference to the accompanying drawings:

[0041] Example:

[0042] like Figure 1 , Figure 2 as well as Figure 3 As shown, the present invention provides a sewing machine thread breakage detection control and emergency response system, including: a thread breakage detection module, including a photoelectric sensor, a tension sensor, a Hall speed sensor and a vision recognition unit, for real-time acquisition and comprehensive recognition of the existence state, tension change, motion continuity and sewing thread shape of the sewing thread;

[0043] Furthermore, the operation procedure of the disconnection detection module includes:

[0044] Photoelectric sensors are used to detect whether the sewing thread is blocking light and output a presence signal;

[0045] Tension sensors are used to detect the working tension of sutures and output tension change signals;

[0046] Hall effect speed sensors are used to acquire the rotational speed of the roller driven by the sewing thread and output a continuous motion signal;

[0047] The visual recognition unit identifies suture trajectory, loose state, disconnection state, and micro-damage state based on images captured by the camera;

[0048] The wire breakage detection module fuses and judges the above signals. When any signal meets the abnormal judgment condition or the recognition result is a wire breakage / micro-damage state, the corresponding judgment signal is output.

[0049] Specifically:

[0050] Signal acquisition:

[0051] The photoelectric sensor is installed above the wire transmission path, with the transmitter and receiver symmetrically arranged. The light beam passes perpendicularly through the sew line trajectory. When the sew line blocks the light path, it outputs a high-level normal signal. When the sew line is broken or the light path is removed, the light path is open and outputs a low-level abnormal signal.

[0052] The tension sensor is connected in series in the suture transmission mechanism near the downstream of the suture pulley. It measures the suture tension change in real time with a range of 0-5N and a sensitivity of 0.01N. The tension data is input as a 0-5V analog voltage signal to the controller.

[0053] The Hall sensor is installed on the end of the roller shaft on the suture transmission path to detect the change in the rotation speed of the roller. When the rotation speed is lower than the third preset threshold of 50 r / min, it is determined that the suture movement is abnormal.

[0054] The visual recognition unit uses a built-in high-definition industrial camera with a frame rate of ≥60FPS and a pixel count of ≥1280×720 to continuously acquire dynamic images of the wire. It also activates a supplementary lighting module to eliminate ambient light interference. The unit identifies the slack and broken state of the suture through image segmentation and edge detection, and identifies the micro-damage state through deep learning algorithms.

[0055] This can be understood as follows: A built-in high-definition industrial camera continuously acquires images of the suture's trajectory, while a supplementary lighting module is simultaneously activated to eliminate ambient light interference, obtaining clear dynamic images of the suture. The acquired images undergo grayscale conversion, Gaussian filtering for noise reduction, and edge enhancement. A threshold segmentation algorithm separates the suture from the background, extracting effective pixels from the suture area. A contour detection algorithm identifies the continuous shape of the suture. If a suture contour interruption is detected with an interruption length ≥ 2mm, it is determined to be a broken suture, and a broken suture identification signal is output. A deep learning semantic segmentation algorithm extracts the surface texture features of the suture, calculates fiber breakage density, surface roughness, and wire diameter change rate, identifies micro-damage states, and classifies them into Level 1, Level 2, and Level 3 micro-damage. The broken suture identification signal or micro-damage classification signal is transmitted to the control and action execution module and simultaneously stored in the data storage and networking module.

[0056] Data processing and fusion:

[0057] The controller performs noise reduction processing on the signals acquired in step a, including: multiple sampling method, Gaussian filtering, threshold comparison and algorithm determination;

[0058] The judgment strategy adopted is as follows:

[0059] If the change in the state of the photoelectric sensor lasts for ≥10ms or the tension output of the tension sensor is <0.3N or higher than the second preset threshold of 3N, or the Hall rotation speed is <50r / min or the visual recognition result shows that the wire breakage length is ≥2mm, or the visual recognition result is a first-level, second-level, or third-level micro-damage;

[0060] The system outputs a corresponding judgment signal based on the signal type: disconnection identification signal / Level 1 micro-damage signal / Level 2 micro-damage signal / Level 3 micro-damage signal.

[0061] The controller generates a corresponding judgment signal and outputs it to the control and action execution module for stopping, speed adjustment, alarm or subsequent processing.

[0062] This can be understood as: the controller determines the type of received signal, distinguishing between disconnection identification signal, first-level micro-damage signal, second-level micro-damage signal, and third-level micro-damage signal;

[0063] If the signal is determined to be a disconnection, immediately perform the following three operations:

[0064] Output a stop control command to the motor drive unit to cut off the power supply to the main motor of the sewing machine;

[0065] Output a braking command to the braking execution unit to lock and brake the sewing machine needle bar mechanism;

[0066] Output continuous audible and visual alarm signals to the alarm output unit to trigger continuous alarm prompts;

[0067] If the signal is determined to be a level one micro-damage signal, only a prompt will be output to the display unit of the human-machine interaction module, displaying "Slight wear on the thread, it is recommended to replace it later", without changing the current operating parameters of the sewing machine and maintaining normal working status;

[0068] If the signal is determined to be a level two micro-damage signal, perform the following two operations in parallel:

[0069] Output speed control command to the motor drive unit to reduce the sewing machine speed to half of the original rated speed;

[0070] It outputs intermittent audible and visual alerts to the alarm output unit, triggering periodic alarm alerts;

[0071] If the signal is determined to be a level three micro-damage signal, proceed as follows:

[0072] Immediately and in parallel output the stop control command to the motor drive unit and the braking command to the braking execution unit;

[0073] The device remains in standby mode, waiting for the operator to complete the suture replacement operation. Once the suture replacement is confirmed to be completed via the visual recognition unit or input unit, the standby mode is deactivated.

[0074] After receiving the reset operation signal transmitted by the input unit of the human-machine interaction module, it outputs a restart command to the system;

[0075] After the system restarts, the comprehensive suture status detection subunit is triggered to re-execute the detection steps S1 to S5, completing a new round of suture status identification.

[0076] Specifically, on an industrial flat-seam production line in a garment factory, real-time detection of upper / lower thread breakage and micro-damage to the sewing thread is performed, and graded responses are implemented to prevent finished products from being scrapped. The frequency of thread breakage and micro-damage is counted to track sewing thread quality or equipment wear issues.

[0077] Scene content:

[0078] Model: Industrial flatbed sewing machine;

[0079] Production cycle time: 60-3000 stitches per minute per machine;

[0080] Fabric: Medium-thickness cotton or polyester fabric (thickness 0.5-2mm);

[0081] Thread: Polyester thread (0.2-0.5mm in diameter);

[0082] Objectives: Achieve a recall rate of ≥95%, a false alarm rate of ≤3%, and an average detection latency of ≤200ms for "broken sutures"; and achieve a classification and identification accuracy of ≥90% for "micro-damage to sutures".

[0083] Experimental objective:

[0084] Verify the accuracy of the wire breakage detection module in identifying wire breaks and micro-damage under real production conditions;

[0085] Measurement and detection delay (the time from the occurrence of a broken wire / micro-damage to the module issuing a judgment signal);

[0086] Compare the impact of different sensor combinations (single photoelectric, photoelectric + tension, photoelectric + vision, full fusion) on detection performance;

[0087] Verify the system's resistance to interference (e.g., suture vibration, ambient light changes, speed changes);

[0088] Verify the effectiveness of the graded response mechanism (speed adjustment, alarm, and shutdown actions corresponding to different micro-damage levels).

[0089] Equipment list:

[0090] One sewing machine was tested.

[0091] Infrared light-shielding photoelectric sensor with a response time ≤1ms;

[0092] Tension sensor, measuring range 0-5N, sensitivity 0.01N;

[0093] Hall effect sensor, resolution ≥ 1° / pulse;

[0094] A camera (industrial camera, ≥60FPS, ≥1280×720 pixels) is connected to an embedded deep learning semantic segmentation and recognition unit;

[0095] Fill light module (white LED ring fill light, brightness adjustable);

[0096] Controller (STM32H743 microcontroller), including high-precision timestamps (ms level);

[0097] Motor drive unit (relay module + PWM speed control module);

[0098] Braking actuator (electromagnetic brake clutch, voltage 24V);

[0099] Alarm output unit (LED indicator light + buzzer);

[0100] Human-computer interaction module (3.5-inch TFT touch screen + 4 mechanical buttons);

[0101] Data storage and networking module (16GBeMMC flash memory + RS485 / WiFi / 4G / 5G communication module);

[0102] Data logging equipment (PC + database / CSV log);

[0103] A controllable wire disconnection device (electromagnetic clamp, response time ≤10ms) is used to trigger wire disconnection at a known time point;

[0104] Suture micro-damage simulation device (sandpaper friction creates different levels of micro-damage);

[0105] The audio / visual and automatic shutdown interfaces are used to verify the response actions.

[0106] Experimental procedure:

[0107] Install photoelectric, tension, Hall sensors and camera / light module. The photoelectric sensor is installed in the middle of the track, the tension sensor is installed at the pull point, the Hall sensor is installed on the wheel shaft, and the camera is aligned with the middle of the track, 5-8cm away from the detection point.

[0108] Baseline data for 60 seconds was collected under conditions of no wire breakage / no micro-damage, and photoelectric level, mean and variance of tension, normal speed fluctuation range, and visual feature template were calculated.

[0109] Threshold settings: lower tension limit = 0.3N, upper tension limit = 3N, lower speed limit = 50r / min, breakage length ≥ 2mm, micro-damage grading threshold (fiber breakage density, Ra value, wire diameter change rate);

[0110] Set the time synchronization and log format, start the sewing machine, and put it into normal sewing mode;

[0111] At preset time points, artificially induce wire breakage, first-level micro-damage, second-level micro-damage, third-level micro-damage, or maintain a normal state as a control;

[0112] The wire breakage detection module collects photoelectric, tension, Hall effect, and visual signals, makes judgments, and outputs corresponding judgment signals;

[0113] Record whether the system issues a judgment signal, the judgment time, the actual status, and the response actions of the control and action execution module;

[0114] Restore the system and proceed to the next test cycle;

[0115] The recorded information includes: whether a disconnection / minor damage occurred, the level of damage; whether it was detected; the detection trigger time or detection delay; and the execution status of response actions (stop / speed adjustment / alarm).

[0116] The control and motion execution module has its input terminal electrically connected to the output terminal of the thread breakage detection module. It is used to receive the thread breakage judgment signal and output the stop control command, braking command or alarm command to realize the automatic control and emergency execution of the sewing machine's operating status.

[0117] Furthermore, the operation flow of the control and action execution module includes:

[0118] The control and motion execution module is electrically connected to the thread breakage detection module. The control and motion execution module includes a controller, a motor drive unit, a braking execution unit, and an alarm output unit. The controller is electrically connected to the motor drive unit, the braking execution unit, and the alarm output unit, respectively. The controller is used to perform logical processing and state decision on the judgment signal. The motor drive unit is used to control the sewing machine main motor to stop or decelerate. The braking execution unit is used to brake or lock the sewing machine needle bar mechanism when thread breakage / level 3 micro-damage occurs. The alarm output unit is used to execute audible and visual alarms or status prompts.

[0119] Specifically:

[0120] The controller of the control and action execution module can be any control device, such as a microcontroller, PLC, DSP or industrial control chip, without being limited to a specific model. The controller is connected to the disconnection detection module and enters the corresponding response logic when it receives a judgment signal.

[0121] The controller can determine the status based on the type of input signal, including: wire breakage identification signal, first-level micro-damage signal, second-level micro-damage signal, and third-level micro-damage signal. First-level micro-damage is when the fiber breakage density is less than 5 fibers / mm², second-level micro-damage is when the fiber breakage density is less than or equal to 5 fibers / mm² and less than 15 fibers / mm², and third-level micro-damage is when the fiber breakage density is greater than or equal to 15 fibers / mm².

[0122] Before sending control commands, the signal is de-jittered and stabilized (continuously monitored for 10ms to confirm signal stability) to prevent false triggering;

[0123] If the signal is a disconnection detection signal: a stop command is sent to the motor drive unit to cut off the main motor power (response delay ≤ 50ms); a braking command is sent to the brake execution unit to drive the electromagnetic brake clutch locking pin mechanism (braking time ≤ 100ms); a continuous audible and visual alarm signal is sent to the alarm output unit (LED red light is constantly on, buzzer sounds continuously at 1kHz).

[0124] If it is a Level 1 micro-damage signal: only send prompt instructions to the human-machine interaction module, do not change the operating parameters, and maintain normal working status;

[0125] If it is a level 2 micro-damage signal: send a speed adjustment command to the motor drive unit to reduce the speed to half of the original rated speed; send an intermittent audible and visual warning signal to the alarm output unit (flashing yellow LED and intermittent buzzer).

[0126] If it is a level 3 micro-damage signal: send a stop command to the motor drive unit and a braking command to the braking execution unit. After the suture replacement is completed and confirmed, receive a reset signal to restart the system and re-detect.

[0127] Braking methods include: electromechanical braking; friction braking; electromagnetic braking; clutch disengagement.

[0128] By setting up control and action execution modules, a graded response to thread breakage and micro-damage is achieved. This not only avoids the scrapping of finished products due to thread breakage, but also reduces the risk of potential failures through micro-damage early warning, improves the quality stability of continuous sewing, and reduces the occurrence of manual inspection or delayed detection of faults.

[0129] The human-computer interaction module is used to display and prompt the thread breakage status, fault type, and sewing machine operating status in real time, and to receive manual operation input;

[0130] Furthermore, the operation process of the human-computer interaction module includes:

[0131] The human-computer interaction module includes a display unit and an input unit;

[0132] The display unit is used to display the location of the wire break, the type of wire break, the level of minor damage, or the fault status information; the input unit is used to receive manual input to perform reset, power on / off, or mode switching operations.

[0133] The human-machine interaction module can be installed on the sewing machine control panel, operation panel or external display unit. The controller is connected to the human-machine interaction module through a communication interface to realize operation information input and status feedback.

[0134] The display unit can be an LCD screen, LED digital screen, touch screen, or indicator light array, without limiting the specific structure. The display content includes: whether a thread break / minor damage has occurred; the type of thread breakage during upper or lower thread insertion; the level of minor damage (Level 1 / Level 2 / Level 3); the location of the fault; the current operating status of the sewing machine (normal / stopped / reduced speed), the current speed, the thread tension value, the frequency statistics of minor damage, the reason for the stop or alarm, automatic recovery prompts or maintenance prompts. The controller can periodically update the display information and highlight the fault with different colors or flashing status.

[0135] The input unit can be a touch button, mechanical button, touch screen, or knob, including: a start button; a stop button; a reset button; a mode switch button; and a menu setting button.

[0136] The input unit is connected to the controller, allowing the operator to perform the following actions after a thread breakage / micro-damage occurs: manual confirmation; fault reset; resumption of sewing; switching sewing modes or speed adjustment; parameter settings (adjusting thresholds and grading standards). The priority of input buttons can be preset, and when a thread breakage or level three micro-damage occurs, the stop and braking operations will be performed first.

[0137] Specifically, the human-computer interaction module's operation process includes: acquiring a judgment signal; displaying the corresponding status information and triggering an alarm; waiting for manual processing or operation input; receiving a reset / start command; and controlling the system to resume operation or enter a re-detection state.

[0138] By setting up a human-machine interaction module, the system enables a visual display of thread breakage and micro-damage status, allowing operators to quickly locate problems and perform recovery operations. This improves the operability and troubleshooting efficiency of the sewing machine, while reducing production losses caused by failure to detect faults in a timely manner.

[0139] The data storage and networking module is used to record and store the thread breakage status, micro-damage level, sensor data, operation records and sewing machine operating status, and upload the data to external devices or network systems through the communication interface to achieve remote monitoring and data management.

[0140] Furthermore, the operation process of the data storage and networking module includes:

[0141] It communicates bidirectionally with the thread breakage detection module and the control and action execution module respectively, and is used to record and store the thread breakage status, sensor data, operation records and sewing machine operating status. It also uploads the data to external devices or network systems through the communication interface to realize remote monitoring and data management.

[0142] Specifically:

[0143] The storage unit can be EEPROM, FLASH, SD card, or industrial storage (preferably 16GBeMMC flash memory in this embodiment); used to store wire breakage detection information (original sensor signal, wire breakage judgment result, response action record); micro-damage related data (image data, micro-damage grading result, fiber breakage density, and other characteristic parameters); cumulative number of faults, control commands and timestamps, historical operating status (speed, tension average), alarm records and reset records; algorithm model parameters (training model parameters of deep learning semantic segmentation algorithm); data is classified and stored according to timestamp (format: YYYY-MM-DDHH:MM:SS.ms) and detection type (wire breakage / micro-damage / normal operation), and can be automatically written according to a preset period, or the event can be recorded immediately after the fault occurs, and important records can be overwritten or stored cyclically, supporting at least 30 days of continuous production data storage;

[0144] The communication unit includes a wired communication interface (RS485, baud rate adjustable from 9600 to 115200bps) and a wireless communication module (WiFi module supports 802.11b / g / n, 4G / 5G module supports full network compatibility). The communication unit is used to send data and status information to external devices and receive control commands and algorithm model update packages sent by the host computer or user terminal.

[0145] The communication methods may include any or a combination of the following: uploading fault events, line break information, and micro-damage data to the host computer; sending operating data to the production line management system; receiving model update packages issued by the remote management system, supporting online model updates and local self-calibration (self-calibration cycle: once every 24 hours); and supporting remote diagnostic or maintenance functions.

[0146] The data storage and networking module can achieve data management in the following ways: real-time data upload (immediate upload in case of failure); historical data reading; remote operation or remote upgrade; fault diagnosis, alarm push and equipment status feedback;

[0147] When a broken wire or minor damage is detected, the module can immediately write the corresponding event, type and related data into the storage area, and send alarm information to the host computer or display device;

[0148] Specifically, the implementation process is as follows: collect data output from the disconnection detection module and the control module → write the running data into the storage unit → if the communication module is enabled, perform data upload or synchronization → display or analyze the data (including micro-damage frequency statistics and disconnection cause correlation analysis) on the host computer or network system → execute control, recovery logic or model update according to external instructions;

[0149] By setting up data storage and networking modules, information on thread breakage and minor damage, as well as sewing status data, can be saved and transmitted in real time, enabling the tracing of fault events, long-term operational status analysis, and remote monitoring. This improves the level of equipment informatization and facilitates the optimization and maintenance scheduling of the production management system.

[0150] Data acquisition and storage process:

[0151] Real-time acquisition: The results of each detection by the disconnection detection module (sampling frequency 500Hz, visual 60FPS) are sent to the data storage and networking module via the data bus;

[0152] Data processing: Denoising and normalization of photoelectric, tension, Hall effect, and visual signals;

[0153] The judgment result is linked to the original sensor value, image data, and processing result with a timestamp to form a complete event record;

[0154] Local storage: Data is written to local storage media in a circular storage mode, supporting at least 30 days of continuous production data storage;

[0155] Example table of stored content (CSV format):

[0156]

[0157] Average data upload latency: ≈330ms;

[0158] The data consistency between local storage and uploaded data is 100%.

[0159] The retransmission function after network disconnection was successfully implemented, and no data was lost in disconnection or minor damage events.

[0160] Experimental results show that the data storage and networking module can achieve high-frequency, multi-sensor real-time acquisition, storage, and reliable uploading of data in a high-speed sewing machine environment, ensuring complete recording of thread breakage and micro-damage events and remote monitoring capabilities of the system.

[0161] Overall workflow: Before the equipment starts running, the thread is first led out by the spool and enters the wire guide assembly set on one side of the frame. The wire guide assembly consists of a wire guide groove, a thread hole, and at least one rotatable guide roller. Under the limiting action of the wire guide groove and the thread hole, the thread enters the detection area in a fixed spatial position. The guide roller rotates by bearing and contacts the thread to reduce running resistance and maintain the continuity of the thread movement. Subsequently, the thread passes through the detection areas of the photoelectric detection unit, tension detection unit, speed detection unit, and vision detection unit in sequence.

[0162] The photoelectric detection unit is fixed on the frame or detection bracket. Its transmitter and receiver are located on both sides of the suture running path. The beam passes horizontally through the space where the suture is located and forms a stable detection optical path. When the suture is present, the photoelectric detection unit outputs a high level, which is the first output state. When the suture breaks, leaves the detection position, or the operation is abnormal, the optical path state changes and outputs a low level, which is the second output state.

[0163] The tension detection unit includes a tension roller that contacts the suture, an elastic clamping mechanism, and a force signal acquisition element. The force signal acquisition element uses a strain gauge sensor. The tension roller is pressed against the suture under the action of elastic preload. The tension changes generated during the operation of the suture are transmitted to the force signal acquisition element through the roller and the clamping mechanism, and converted into a 0-5V analog electrical signal output. When the tension value is lower than 0.3N or higher than 3N, an abnormal tension signal is triggered.

[0164] The rotation speed detection unit is set on the rotating shaft of the wheel that moves synchronously with the suture. A magnet is arranged every 90° on the rotating shaft, and a Hall sensor is fixed at the corresponding position of the magnet. During the rotation of the rotating shaft, the Hall sensor outputs a pulse signal. The frequency of the pulse signal is positively correlated with the rotation speed of the wheel. When the rotation speed is lower than 50 r / min, it is determined that the suture movement is abnormal.

[0165] The visual inspection unit includes a camera component and a ring-shaped supplementary lighting component fixed diagonally above the inspection area. The camera component is at a 45° angle to the suture. The camera component continuously acquires images of the spatial area where the suture is located at a frequency of 60 FPS. The acquired image signals are transmitted to the suture status comprehensive inspection subunit via the SPI interface for analysis and processing. The subunit sequentially performs grayscale conversion, filtering, edge detection, and semantic segmentation to identify whether the suture is interrupted. If the interruption length is ≥2mm, it is determined to be a broken suture. At the same time, it identifies whether there is micro-damage to the suture. Micro-damage is divided into first-level, second-level, and third-level.

[0166] Photoelectric detection signals, tension detection signals, speed detection signals, and visual detection image data are all input to the controller of the control and motion execution module via wired connection. The wired connection uses a GPIO interface or an ADC interface. After the system is powered on, the controller first executes the initialization process to confirm the communication status, reference signals, and operating parameters of each detection unit. The reference signals include the photoelectric high-level threshold, tension baseline value, and speed baseline frequency. The operating parameters include a sampling frequency of 500Hz and a de-jitter time of 10ms. During the operation of the sewing machine, the controller continuously collects the real-time data output by each detection unit according to the preset sampling period.

[0167] The collected data undergoes Gaussian filtering, jitter removal, and time synchronization processing before entering the judgment process. The Gaussian filtering kernel size is 5×5, jitter removal requires three consecutive sampling signals to be consistent, and time synchronization involves binding millisecond-level timestamps. The controller performs joint analysis on the state changes of the photoelectric detection signal, the numerical changes of the tension signal, the frequency changes of the rotation speed pulse, and the visual detection results according to preset logic. A corresponding judgment signal is generated when any of the following conditions are met:

[0168] The photoelectric signal remains low for ≥10ms;

[0169] Tension signal <0.3N or >3N;

[0170] The rotational speed pulse frequency corresponds to a rotational speed < 50 r / min;

[0171] Visual recognition of suture breaks ≥2mm will generate a suture breakage identification signal.

[0172] Visually identified sutures as having level one micro-damage, with a fiber breakage density of <5 fibers / mm².

[0173] Visually identified sutures as having secondary micro-damage, with 5 / mm² ≤ fiber breakage density < 15 / mm²;

[0174] Visually identified sutures as having level three micro-damage, with a fiber breakage density ≥15 fibers / mm².

[0175] After generating the judgment signal, the controller executes the corresponding action according to the preset control flow:

[0176] If the signal is a broken thread or a level 3 micro-damage signal: immediately send a stop control command to the sewing machine drive system to cut off the main motor power, with a response delay of ≤50ms; simultaneously output a 24V voltage signal to the braking mechanism to drive the electromagnetic brake clutch to lock and brake the sewing machine spindle and needle bar, with a braking time of ≤100ms; synchronously output status information to the human-machine interface module, including the broken thread or level 3 micro-damage and the fault location, triggering a constantly lit red LED and a continuous alarm buzzer; and write the event data to the data storage module, including the timestamp, original sensor values, and judgment result.

[0177] If it is a level 2 micro-damage signal: a speed adjustment command is sent to the drive system to reduce the sewing machine speed to half of the original rated speed through the PWM signal; status information, i.e., level 2 micro-damage, is output to the human-machine interface module, triggering the yellow LED to flash and the buzzer to sound an intermittent alarm; event data is stored synchronously.

[0178] If it is a Level 1 micro-damage signal: only output a prompt message to the human-machine interaction module, namely that the suture is slightly worn and it is recommended to replace it later, maintain the current operating parameters, and store the micro-damage data;

[0179] After the fault is resolved, when the operator re-threads the cable and inputs a reset command via the reset button on the human-machine interface module, the controller re-enters the detection and confirmation process to re-collect and judge the output status of each detection unit:

[0180] The photoelectric detection unit outputs a high level.

[0181] The output value of the tension detection unit is in the range of 0.5-2N, which is the normal tension range;

[0182] The speed detection signal corresponds to a speed ≥100 r / min;

[0183] The visual inspection unit failed to detect broken wires and micro-damage;

[0184] On the other hand, it includes a processor, a memory, and a computer program stored in the memory, wherein the processor executes the computer program and applies it to the above-described sewing machine thread breakage detection control and emergency response system.

[0185] After all the above conditions are met, the controller sends a release control command to the braking mechanism, cuts off the 24V voltage, and controls the drive system to start the sewing machine in low-speed operation mode, which is 500 stitches / minute. During low-speed operation, the control unit continues to collect and analyze various detection signals at a frequency of 500Hz. After 3 seconds without any abnormalities, the machine returns to the original set operating state according to the control process. The original set operating state includes speed and production cycle, thus completing the complete implementation process of sewing machine thread breakage and micro-damage detection, control processing, and resumption of operation.

[0186] The above embodiments are only used to illustrate the technical solutions of the present invention, and are not intended to limit it. Although the present invention has been described in detail with reference to the foregoing embodiments, those skilled in the art should understand that modifications can still be made to the technical solutions described in the foregoing embodiments, or equivalent substitutions can be made to some of the technical features. Such modifications or substitutions will not cause the essence of the corresponding technical solutions to deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. A sewing machine thread breakage detection, control, and emergency response system, characterized in that, The system includes: The thread breakage detection module includes a photoelectric sensor, a tension sensor, a Hall speed sensor, and a vision recognition unit, which is used to collect and comprehensively identify the presence, tension changes, motion continuity, and morphology of the sewing thread in real time. The control and action execution module has its input terminal electrically connected to the output terminal of the thread breakage detection module. It is used to receive the thread breakage judgment signal and output the stop control command, braking command or alarm command to realize the automatic control and emergency execution of the sewing machine's operating status. The human-computer interaction module is used to display and prompt the thread breakage status, fault type, and sewing machine operating status in real time, and to receive manual operation input; The data storage and networking module communicates bidirectionally with the thread breakage detection module and the control and action execution module, respectively, to record and store the thread breakage status, sensor data, operation records and sewing machine operating status, and uploads the data to external devices or network systems through the communication interface to realize remote monitoring and data management; The wire breakage detection module fuses the signals from each sensor with the visual recognition results to make a judgment, and outputs a wire breakage judgment signal when the preset wire breakage conditions are met. The visual recognition unit is also used for suture breakage identification and suture micro-damage grading. The visual recognition unit includes: The image acquisition subunit is used to continuously acquire images of the sewing thread's trajectory using a built-in high-definition industrial camera. The supplementary light subunit is used to eliminate ambient light interference and obtain clear dynamic images of the sutures; The image preprocessing subunit is used to perform grayscale conversion, Gaussian filtering for noise reduction, and edge enhancement on the acquired image. The image segmentation subunit is used to separate the suture from the background and extract the effective pixels of the suture region using a threshold segmentation algorithm. The contour detection subunit is used to identify the continuous shape of the suture. When the suture contour is interrupted and the interruption length is ≥2mm, it outputs a suture break recognition signal. The deep learning processing subunit is used to perform pixel-level segmentation of the suture region using a semantic segmentation algorithm, extract feature parameters, and identify the micro-damage state based on the feature parameters and classify it into first-level micro-damage, second-level micro-damage, or third-level micro-damage. The feature parameters include fiber breakage density, surface roughness, and wire diameter change rate.

2. The sewing machine thread breakage detection, control, and emergency response system according to claim 1, characterized in that, The photoelectric sensor is positioned above the sewing thread transmission path to detect whether the sewing thread is blocking light and output a presence signal. The tension sensor is connected in series in the suture transmission mechanism to detect the working tension of the suture and output a tension change signal; The Hall speed sensor is installed on the end of the wheel shaft on the suture transmission path to collect the wheel speed for characterization. When the speed is lower than the third preset threshold, it is determined that the suture movement is interrupted.

3. The sewing machine thread breakage detection, control, and emergency response system according to claim 2, characterized in that, In the broken thread detection module, the visual recognition unit identifies the suture trajectory, slack state, and breakage state based on images captured by the camera. The disconnection detection module performs fusion judgment on the above signals, and outputs the disconnection judgment result when any signal meets the abnormal judgment condition or the recognition result is a disconnection state.

4. The sewing machine thread breakage detection, control, and emergency response system according to claim 3, characterized in that, The control and action execution module is electrically connected to the disconnection detection module. The control and action execution module includes a controller, a motor drive unit, a braking execution unit, and an alarm output unit. The controller is electrically connected to the motor drive unit, the braking execution unit, and the alarm output unit, respectively.

5. The sewing machine thread breakage detection, control, and emergency response system according to claim 4, characterized in that, In the control and action execution module, the controller is used to perform logical processing and state decision on the thread breakage judgment signal, the motor drive unit is used to control the sewing machine main motor to stop and decelerate, the braking execution unit is used to brake or lock the sewing machine needle bar mechanism when a thread breakage occurs, the alarm output unit is used to execute audible and visual alarms or status prompts, and the human-machine interaction module includes a display unit and an input unit. The display unit is used to display the thread breakage location, thread breakage type, and fault status information, and the input unit is used to receive manual input to perform reset, power on / off, and mode switching operations.

6. The sewing machine thread breakage detection, control, and emergency response system according to claim 1, characterized in that, The data storage and networking module includes a storage unit and a communication unit. The storage unit is used to store disconnection detection data, fault records, and historical operating parameters. The communication unit includes a wired communication interface and a wireless communication interface.

7. The sewing machine thread breakage detection, control, and emergency response system according to claim 6, characterized in that, The storage unit is used to store the image data, micro-damage grading data and processing results of each step output by the visual recognition unit, and is classified and stored according to the format of timestamp and detection type; the storage unit is also used to pre-store the training model parameters of the deep learning semantic segmentation algorithm, support online model updates and local self-calibration, and receive update packets through the communication unit and automatically complete calibration; The communication unit includes an RS485 wired communication interface, a WiFi wireless communication interface, and a 4G / 5G communication module, which supports uploading suture micro-damage data, suture breakage detection data, statistical analysis reports, and micro-damage frequency to the remote management system.

8. A sewing machine thread breakage detection, control, and emergency response system according to claim 5, characterized in that, The visual recognition unit is also used for early warning and control based on the suture micro-damage grading, and the visual recognition unit includes: The first-level micro-damage control subunit is used to send a prompt command to the human-machine interaction module of the motor drive unit when the suture micro-damage is classified as first-level micro-damage, without changing the operating parameters and maintaining the normal working state. The secondary micro-damage control subunit is used to output a speed control command to the motor drive unit when the suture micro-damage is classified as secondary micro-damage, reducing the sewing machine speed to half of the original rated speed; and to output intermittent audible and visual prompts to the alarm output unit to trigger periodic alarm prompts. The Level 3 Micro-damage Control Subunit is used to output a stop control command to the motor drive unit and a braking command to the braking execution unit when the suture micro-damage is classified as Level 3 micro-damage.

9. A sewing machine thread breakage detection and control computer device, comprising a processor, a memory, and a computer program stored in the memory, characterized in that, When the processor executes the computer program, it implements the sewing machine thread breakage detection, control and emergency response system as described in any one of claims 1-8.

Images