Full-automatic intelligent bobbin winder based on multi-shaft cooperative control

Abstract

The invention discloses a full-automatic intelligent bobbin winder based on multi-shaft cooperative control, and belongs to the field of textile machinery, and the full-automatic intelligent bobbin winder comprises a multi-shaft driving module used for carrying out constant-line-speed and constant-tension winding on yarns; the tension dynamic compensation module is used for monitoring the yarn tension in real time and adjusting the Y-axis rotating speed and the pneumatic compensation amount in the multi-axis driving module in real time through a PID algorithm based on a mathematical model of yarn tension-speed-winding diameter; the intelligent detection module is used for capturing a yarn surface image, analyzing the yarn surface image and dynamically adjusting winding tension or triggering a yarn suction arm to remove based on an analysis result; the broken yarn self-repairing module is used for pre-judging the broken yarn risk 0.5 second in advance, positioning the broken yarn position and automatically repairing the yarn; and the control system is used for cooperatively controlling the multi-axis driving module, the tension dynamic compensation module, the intelligent detection module and the broken yarn self-repairing module.

Description

Technical Field

[0001] The invention belongs to the field of textile machinery, and in particular relates to a fully automatic intelligent winding machine based on multi-axis collaborative control. Background Art

[0002] The winding machine is a special equipment in the textile industry. It is used to wind the yarn from the bobbin or hank into a cone. It is the last process of spinning and the first process of weaving. It plays a "bridge" role between the upper and lower processes. Its main function is to connect the bobbins or hanks with smaller capacity and wind them into cones with larger capacity to improve the production efficiency of subsequent processes. The yarn clearing device is used to check the yarn, remove defects and impurities such as thick sections, thin sections, double yarns, weak twist yarns, cotton knots, etc., to improve the yarn quality. Ensure that the cone yarn is well wound and formed, which is convenient for unwinding in subsequent processes.

[0003] Traditional winding machines have the following technical pain points: the mechanical tensioner cannot adapt to changes in yarn type / speed, which easily causes uneven winding density. Manual yarn splicing takes more than 30 seconds per time, affecting equipment efficiency. At the same time, the photoelectric detection recognition rate for micron-level hairiness / foreign fibers is less than 70%. Summary of the invention

[0004] In order to solve the above technical problems, the present invention provides a fully automatic intelligent winding machine based on multi-axis collaborative control, comprising:

[0005] Multi-axis drive module, used for winding the yarn at constant speed and constant tension;

[0006] A tension dynamic compensation module is used to monitor the yarn tension in real time, and based on the mathematical model of yarn tension-speed-winding diameter, adjust the Y-axis speed and pneumatic compensation amount in the multi-axis drive module in real time through a PID algorithm;

[0007] An intelligent detection module is used to capture the yarn surface image, analyze the yarn surface image, and dynamically adjust the winding tension or trigger the yarn suction arm to reject based on the analysis result;

[0008] The yarn break self-repair module is used to predict the risk of yarn breakage 0.5 seconds in advance, locate the yarn breakage position, and automatically complete the yarn repair;

[0009] A control system is used to coordinately control the multi-axis driving module, the tension dynamic compensation module, the intelligent detection module and the yarn breakage self-repair module.

[0010] Preferably, the multi-axis driving module includes an X-axis, a Y-axis and a Z-axis:

[0011] The X-axis is a yarn guide, which is used to guide the yarn to move in the horizontal direction;

[0012] The Y axis is a grooved drum, which is used to control the winding speed and direction of the yarn to ensure uniform tension of the yarn during the winding process;

[0013] The Z axis is a bobbin holder, which is used to firmly clamp the bobbin and adjust the height of the bobbin according to the winding progress to ensure the continuity of the yarn winding.

[0014] Preferably, the tension dynamic compensation module comprises:

[0015] High-precision tension sensor for real-time detection of yarn tension;

[0016] A pneumatic fine-tuning cylinder for dynamically compensating tension fluctuations caused by winding diameter changes or mechanical inertia based on the yarn tension;

[0017] The PID controller is used to calculate the tension compensation amount that needs to be adjusted according to the feedback value of the high-precision tension sensor through a PID algorithm, and output a control signal to the multi-axis drive module.

[0018] Preferably, the PID controller is also used to construct a winding diameter calculation model, calculate the winding diameter in real time based on the winding diameter calculation model, and dynamically adjust the Y-axis rotation speed in the multi-axis driving module based on the winding diameter.

[0019] Preferably, the intelligent detection module comprises:

[0020] High-definition camera, used to collect image data of the yarn surface as the original input for detection;

[0021] Dual light source system to provide stable lighting for HD cameras and enhance image quality;

[0022] An image preprocessing unit, used to preprocess the images captured by the high-definition camera to obtain a preprocessed data set;

[0023] A deep learning unit is used to analyze the preprocessed dataset based on the improved YOLOv5 model to identify and classify yarn defects.

[0024] Preferably, the improved YOLOv5 model includes:

[0025] An input layer is used to scale the preprocessed dataset to 640×640 and normalize it to [0,1];

[0026] Backbone unit, used to detect minor yarn defects based on CSPDarknet53+ cross-stage local attention module;

[0027] Neck unit, used to fuse multi-scale features based on PANet+dynamic convolution;

[0028] Head unit, used to optimize the anchor frame size and adapt it to the yarn defect size.

[0029] Preferably, the yarn breakage self-repair module comprises:

[0030] Rotating yarn suction arm is used to quickly locate and suck the broken yarn when the yarn breaks;

[0031] Piezoelectric ceramic splicer is used to repair broken yarn by splicing.

[0032] Preferably, the yarn break self-repair module further comprises: an acoustic emission sensor for predicting the risk of yarn breakage in advance so as to reduce the speed in advance and prepare for the repair action.

[0033] Preferably, the multi-axis driving module, the tension dynamic compensation module, the intelligent detection module and the yarn breakage self-repairing module are connected to a control system via an EtherCAT communication protocol.

[0034] Preferably, the control system also includes a human-machine interface for parameter setting, status monitoring and data docking with the MES system.

[0035] Compared with the prior art, the present invention has the following advantages and technical effects:

[0036] 1. Multi-axis collaborative winding: X / Y / Z axis linkage realizes constant line speed and constant tension winding. Compared with traditional equipment, the uniformity of winding density is improved by 40%.

[0037] 2. Pneumatic-algorithm dual-mode compensation: Mechanical compensation (pneumatic) and electronic compensation (PID frequency modulation) work together to adapt to the yarn strength range of 0.1-3.0N / tex.

[0038] 3. Embedded AI edge computing: Deploy a lightweight yarn defect model on the FPGA chip, with a detection delay of <2ms, without the need for an external industrial computer. BRIEF DESCRIPTION OF THE DRAWINGS

[0039] The drawings constituting a part of the present application are used to provide a further understanding of the present application. The illustrative embodiments and descriptions of the present application are used to explain the present application and do not constitute an improper limitation on the present application. In the drawings:

[0040] Figure 1 A multi-axis collaborative control logic diagram of an embodiment of the present invention;

[0041] Figure 2 is a flow chart of a yarn defect detection algorithm according to an embodiment of the present invention;

[0042] Figure 3 The figure is a diagram showing the overall framework of the system according to an embodiment of the present invention. DETAILED DESCRIPTION

[0043] It should be noted that, in the absence of conflict, the embodiments and features in the embodiments of the present application can be combined with each other. The present application will be described in detail below with reference to the accompanying drawings and in combination with the embodiments.

[0044] It should be noted that the steps shown in the flowcharts of the accompanying drawings can be executed in a computer system such as a set of computer executable instructions, and that, although a logical order is shown in the flowcharts, in some cases, the steps shown or described can be executed in an order different from that shown here.

[0045] Embodiment 1

[0046] like Figure 1 , Figure 2 and Figure 3 As shown, in this embodiment, a fully automatic intelligent winding machine based on multi-axis collaborative control is provided, comprising:

[0047] Multi-axis drive module, used for winding the yarn at constant speed and constant tension;

[0048] A tension dynamic compensation module is used to monitor the yarn tension in real time, and based on the mathematical model of yarn tension-speed-winding diameter, adjust the Y-axis speed and pneumatic compensation amount in the multi-axis drive module in real time through a PID algorithm;

[0049] An intelligent detection module is used to capture the yarn surface image, analyze the yarn surface image, and dynamically adjust the winding tension or trigger the yarn suction arm to reject based on the analysis result;

[0050] The yarn break self-repair module is used to predict the risk of yarn breakage 0.5 seconds in advance, locate the yarn breakage position, and automatically complete the yarn repair;

[0051] A control system is used to coordinately control the multi-axis driving module, the tension dynamic compensation module, the intelligent detection module and the yarn breakage self-repair module.

[0052] Multi-axis coordinated control is one of the core means of this technical solution. By accurately coordinating the movement of the X-axis (yarn guide), Y-axis (grooved drum) and Z-axis (bobbin holder), a constant linear speed and constant tension winding effect is achieved. The specific modules include:

[0053] High-precision drive system: The X-axis is driven by a linear motor with a repeatability accuracy of ±5μm; the Y-axis uses a magnetic suspension variable frequency motor with a stepless speed range of 3000-12000rpm; the Z-axis uses a pneumatic adaptive fixture to support bobbins of various specifications.

[0054] X-axis (yarn guide): driven by a linear motor, responsible for guiding the yarn to move horizontally, with a repeatability accuracy of ±5μm. Its function is to ensure that the yarn is evenly distributed during the winding process and avoid yarn accumulation or uneven winding.

[0055] Y axis (grooved drum): driven by a magnetic suspension variable frequency motor, with a speed range of 3000-12000rpm and stepless speed change. There are spiral grooves on the surface of the grooved drum, and the yarn guide angle is 15°, which is used to control the winding speed and direction of the yarn to ensure uniform yarn tension during the winding process.

[0056] Z axis (bobbin holder): It adopts pneumatic adaptive clamp, supports bobbins of Φ50-Φ300mm, and can be lifted and lowered vertically (stroke 500mm). Its function is to firmly clamp the bobbin and adjust the height of the bobbin according to the winding progress to ensure the continuity of yarn winding.

[0057] Multi-axis linkage algorithm: Through the multi-axis collaborative algorithm, the position synchronization of X / Y / Z axes is ensured (interpolation cycle 1ms), constant linear speed winding is achieved, and winding quality problems caused by speed mismatch are avoided.

[0058] Real-time calculation and dynamic adjustment of winding diameter: The winding diameter calculation model is used to update the Y-axis speed in real time to ensure the stability of the tension during the winding process.

[0059] The tension dynamic compensation module includes:

[0060] High-precision tension sensor: range 0-50cN, sampling rate 1kHz, real-time detection of yarn tension.

[0061] Pneumatic fine-tuning cylinder: stroke ±2mm, response time <10ms, used for mechanical compensation.

[0062] PID closed-loop control: Based on the mathematical model of yarn tension-speed-winding diameter, the Y-axis speed and pneumatic compensation are adjusted in real time through the PID algorithm to control the tension fluctuation within the CV value ≤ 1.5%.

[0063] Intelligent detection and yarn defect recognition modules include:

[0064] Machine vision module: It uses a dual-light source (ring LED + infrared backlight) high-definition camera with 5 million pixels and a frame rate of 200fps, which can clearly capture the surface image of the yarn.

[0065] Improved YOLOv5 model: Optimized for yarn defect detection, supports classification and identification of neps, hairiness, and foreign fibers, with a detection rate of >99.5% and a false alarm rate of <0.3%.

[0066] Dynamic compensation and execution linkage: According to the type and position of yarn defects, the winding tension is dynamically adjusted or the suction arm is triggered to remove the yarn to achieve closed-loop control.

[0067] The yarn break self-repair module includes:

[0068] Acoustic emission sensor: frequency range 20-100kHz, predicting the risk of yarn breakage 0.5 seconds in advance, reducing equipment damage when yarn breaks.

[0069] Rotary yarn suction arm and piezoelectric ceramic splicer: yarn suction arm positioning accuracy ±0.1mm, splicing strength> 90% of original yarn, yarn breakage processing time ≤ 3 seconds.

[0070] After detecting a yarn break, the module can automatically repair the yarn without human intervention:

[0071] The rotating yarn suction arm can quickly move to the broken yarn position with a positioning accuracy of ±0.1mm. The vacuum nozzle (vacuum degree -80kPa) at the end of the yarn suction arm can quickly suck the broken yarn to prevent the yarn from loosening. With the piezoelectric ceramic splicer, the broken yarn is spliced ​​and repaired. The splicing strength exceeds 90% of the original yarn, ensuring the reliable strength of the repaired yarn.

[0072] Quick restart: After completing the twisting, the equipment can resume winding in a very short time. The entire broken yarn repair process is controlled within 3 seconds, which significantly improves the operating efficiency of the equipment.

[0073] Embedded AI and edge computing modules include:

[0074] Lightweight model deployment: The improved YOLOv5 model is deployed on the FPGA chip, with a detection delay of <2ms and no external industrial computer required.

[0075] Adaptive optimization: The online learning module can dynamically update the model weights to adapt to different yarn types and enhance the versatility and flexibility of the equipment.

[0076] Embodiment 2

[0077] like Figure 1 As shown, in this embodiment, a fully automatic intelligent winding machine based on multi-axis collaborative control is provided, comprising:

[0078] 1. Equipment structure design

[0079] (1) Core Module

[0080] Multi-axis coordinated drive system

[0081] X-axis (yarn guide): linear motor driven, repeat positioning accuracy ±5μm.

[0082] Y-axis (grooved drum): magnetic suspension variable frequency motor, speed range 3000-12000rpm (stepless speed change).

[0083] Z axis (bobbin holder): pneumatic adaptive clamp, supports bobbins of Φ50-Φ300mm.

[0084] Tension dynamic compensation unit:

[0085] High-precision tension sensor (range 0-50cN, sampling rate 1kHz).

[0086] Pneumatic fine-tuning cylinder (stroke ±2mm, response time <10ms).

[0087] (2) Intelligent detection system

[0088] Machine Vision Module:

[0089] Dual light source (ring LED + infrared backlight) HD camera (5 megapixels, frame rate 200fps).

[0090] Deep learning algorithm: Improved yarn defect detection model based on YOLOv5 (supports classification and identification of neps, hairiness, and foreign fibers).

[0091] (3) Yarn break self-repair mechanism:

[0092] Rotary yarn suction arm (vacuum degree -80kPa), with piezoelectric ceramic splicer (splicing strength> 90% of original yarn).

[0093] Control logic and algorithm:

[0094] (1) Tension closed-loop control:

[0095] Establish a mathematical model of yarn tension-speed-winding diameter:

[0096]

[0097] Wherein, T is the theoretical tension, v is the line speed, D is the current roll diameter, D0 is the initial roll diameter, K1 / K2 is the material coefficient, and μ is the friction factor.

[0098] The Y-axis speed and pneumatic compensation are adjusted in real time through the PID algorithm, and the tension fluctuation CV value is ≤1.5%.

[0099] (2) Yarn defect identification and classification:

[0100] A training set of 100,000 yarn defect images was constructed, and the model was optimized using transfer learning. The detection rate was >99.5% and the false alarm rate was <0.3%.

[0101] (3) Quick response to yarn breakage:

[0102] Based on the acoustic emission sensor (frequency range 20-100kHz), the yarn risk is pre-judged and the speed is reduced 0.5s in advance. The positioning accuracy of the yarn suction arm is ±0.1mm.

[0103] The operation process includes:

[0104] 1. Equipment Assembly

[0105] The multi-axis drive system, pneumatic compensation unit and vision module are integrated into the aluminum alloy frame. The overall size of the machine is 2000×1200×1800mm and the weight is ≤1.8 tons.

[0106] 2. Operation test (cotton yarn 40Ne)

[0107] Winding test: At a speed of 1500m / min, the tension fluctuation range is ±0.15cN, and the package CV value is 1.2%.

[0108] Yarn break simulation: After the yarn is cut manually, the yarn suction-twisting-restart process is completed within 2.8 seconds.

[0109] Yarn defect detection: When yarn containing 0.5% foreign fibers is injected, the system detection rate is 99.7% and the false rejection rate is 0.25%.

[0110] The module connection relationship is shown in Table 1:

[0111] Table 1

[0112]

[0113] The detailed processing steps of the yarn defect detection algorithm include:

[0114] 1. Overall algorithm process:

[0115] Input: Yarn surface image (resolution 2592×1944, 8-bit grayscale) captured by a 5-megapixel high-definition camera (200fps).

[0116] Output: yarn defect type (neps, hairiness, foreign fibers), position coordinates, size grade (Classimat 3 classification).

[0117] 2. Step-by-step processing logic

[0118] Step 1: Image data acquisition and preprocessing

[0119] (1) Image acquisition:

[0120] Trigger condition: When the yarn speed is ≥500m / min, the camera will shoot synchronously according to the encoder pulse (trigger once every 0.5m).

[0121] Light source configuration: ring LED (brightness 5000lux) + infrared backlight (wavelength 850nm), to eliminate yarn shaking and blur.

[0122] 2. Preprocessing:

[0123] Denoising: Non-local means filtering (NL-Means, search window 15×15, similarity window 5×5, h=10).

[0124] Enhancement: Adaptive histogram equalization (CLAHE, grid size 32×32, contrast limit 2.0).

[0125] Binarization: Otsu threshold segmentation + morphological opening operation (3×3 cross structure element) to remove small noise.

[0126] Step 2: Yarn defect feature extraction and classification

[0127] 1. Improve YOLOv5 model architecture:

[0128] Input layer: The image is resized to 640×640 and normalized to [0,1].

[0129] Backbone: CSPDarknet53 + Cross-stage Local Attention Module (CSLA) to enhance the ability to detect minor yarn defects.

[0130] Neck: PANet+dynamic convolution (DyHead), integrating multi-scale features.

[0131] Head: The output layer anchor frame size is optimized to (8×8, 16×16, 32×32) to adapt to the yarn defect size (0.1-3mm).

[0132] 2. Training data and parameters:

[0133] Dataset: 100,000 annotated images (50,000 neps, 30,000 hairiness, 20,000 foreign matter), data augmentation (rotation ±15°, Gaussian noise SNR=30dB).

[0134] Training parameters**: Adam optimizer (lr=0.001, cosine annealing), BatchSize=32, Epoch=300.

[0135] 3. Reasoning process:

[0136] Forward computing: FP16 quantization acceleration, single frame inference time ≤ 1.5ms (NVIDIA Jetson AGX Xavier).

[0137] Post-processing: non-maximum suppression (NMS, IoU threshold 0.5), confidence threshold 0.7.

[0138] Step 3: Dynamic compensation and execution linkage

[0139] 1. Yarn defect position mapping:

[0140] According to the yarn speed (v) and the camera trigger time (t), the axial position of the yarn defect on the package is calculated:

[0141] L_{\text{defect}}=v\cdott+\text{encoder accumulated pulse}\cdot\frac{\piD}{N_{\text{encoder resolution}}}

[0142] \]

[0143] 2. Dynamic compensation logic: Minor yarn defects (Classimat 1-2): record the position and reduce the local tension in subsequent winding (PID target value is reduced by 10%).

[0144] 3. Severe yarn defects (Classimat 3 and above): trigger the yarn suction arm to remove the yarn and suspend winding (downtime ≤ 3 seconds).

[0145] Step 4: Result output and closed-loop control

[0146] 1. Data output:

[0147] Yarn defect statistics report: type, quantity, distribution location (JSON format, uploaded to MES system via TCP / IP).

[0148] Real-time control signal: tension compensation (ΔT) → pneumatic fine-tuning cylinder displacement (ΔL = ΔT / K, K = 0.2 cN / μm),

[0149] Yarn breaking coordinates → yarn suction arm motion trajectory (three-axis interpolation control).

[0150] Algorithm technology advantages

[0151] 1. High-precision detection: The detection rate of cotton neps with a diameter of ≥ 0.1 mm is 99.5%, and the foreign fiber recognition rate is 98.7% (compared with 97% of UsterQuantum3).

[0152] 2. Low-latency processing: The end-to-end delay from image acquisition to compensation execution is ≤5ms, meeting the high-speed requirement of 1800m / min.

[0153] 3. Adaptive optimization: The online learning module (based on PyTorchLite) can dynamically update the model weights to adapt to different yarn varieties.

[0154] 4. Control Logic Verification

[0155] 1. Test conditions:

[0156] Yarn type: 40Ne cotton yarn, speed 1500m / min, package diameter Φ200mm.

[0157] Artificially implanted yarn defects: 20 cotton neps (0.3 mm) and 5 foreign fibers (1 mm polyester yarn).

[0158] 2. The test results are shown in Table 2:

[0159] Table 2

[0160]

[0161] This embodiment realizes the intelligent upgrade of the high-speed winding machine through the deep coupling of multi-axis collaborative control logic and embedded yarn defect detection algorithm. The key innovations include:

[0162] 1. Dynamic tension-speed-reel model: accurately decouple mechanical inertia force and friction accumulation effect.

[0163] 2. Lightweight YOLOv5 improved model: sub-millisecond yarn defect detection on FPGA.

[0164] 3. Closed-loop compensation mechanism: The yarn defect classification and processing strategy takes into account both efficiency and quality.

[0165] The above are only preferred specific implementations of the present application, but the protection scope of the present application is not limited thereto. Any changes or substitutions that can be easily thought of by a person skilled in the art within the technical scope disclosed in the present application should be included in the protection scope of the present application. Therefore, the protection scope of the present application should be based on the protection scope of the claims.

Claims

1. A fully automatic intelligent winding machine based on multi-axis collaborative control, characterized in that: include: Multi-axis drive module, used for winding the yarn at constant speed and constant tension; A tension dynamic compensation module is used to monitor the yarn tension in real time, and based on the mathematical model of yarn tension-speed-winding diameter, adjust the Y-axis speed and pneumatic compensation amount in the multi-axis drive module in real time through a PID algorithm; An intelligent detection module is used to capture the yarn surface image, analyze the yarn surface image, and dynamically adjust the winding tension or trigger the yarn suction arm to reject based on the analysis result; The yarn break self-repair module is used to predict the risk of yarn breakage 0.5 seconds in advance, locate the yarn breakage position, and automatically complete the yarn repair; A control system is used to coordinately control the multi-axis driving module, the tension dynamic compensation module, the intelligent detection module and the yarn breakage self-repair module.

2. The winding machine according to claim 1, characterized in that: The multi-axis drive module includes an X-axis, a Y-axis and a Z-axis: The X-axis is a yarn guide, which is used to guide the yarn to move in the horizontal direction; The Y axis is a grooved drum, which is used to control the winding speed and direction of the yarn to ensure uniform tension of the yarn during the winding process; The Z axis is a bobbin holder, which is used to firmly clamp the bobbin and adjust the height of the bobbin according to the winding progress to ensure the continuity of the yarn winding.

3. The winding machine according to claim 1, characterized in that: The tension dynamic compensation module comprises: High-precision tension sensor for real-time detection of yarn tension; A pneumatic fine-tuning cylinder for dynamically compensating tension fluctuations caused by winding diameter changes or mechanical inertia based on the yarn tension; The PID controller is used to calculate the tension compensation amount that needs to be adjusted according to the feedback value of the high-precision tension sensor through a PID algorithm, and output a control signal to the multi-axis drive module.

4. The winding machine according to claim 3, characterized in that: The PID controller is also used to construct a winding diameter calculation model, calculate the winding diameter in real time based on the winding diameter calculation model, and dynamically adjust the Y-axis rotation speed in the multi-axis driving module based on the winding diameter.

5. The winding machine according to claim 1, characterized in that: The intelligent detection module comprises: High-definition camera, used to collect image data of the yarn surface as the original input for detection; Dual light source system to provide stable lighting for HD cameras and enhance image quality; An image preprocessing unit, used to preprocess the images captured by the high-definition camera to obtain a preprocessed data set; A deep learning unit is used to analyze the preprocessed dataset based on the improved YOLOv5 model to identify and classify yarn defects.

6. The winding machine according to claim 5, characterized in that: The improved YOLOv5 model includes: An input layer is used to scale the preprocessed dataset to 640×640 and normalize it to [0,1]; Backbone unit, used to detect minor yarn defects based on CSPDarknet53+ cross-stage local attention module; Neck unit, used to fuse multi-scale features based on PANet+dynamic convolution; Head unit, used to optimize the anchor frame size and adapt it to the yarn defect size.

7. The winding machine according to claim 1, characterized in that: The yarn break self-repair module comprises: Rotating yarn suction arm is used to quickly locate and suck the broken yarn when the yarn breaks; Piezoelectric ceramic splicer is used to repair broken yarn by splicing.

8. The winding machine according to claim 7, characterized in that: The yarn break self-repair module further comprises: an acoustic emission sensor for predicting the risk of yarn breakage in advance so as to reduce the speed in advance and prepare for the repair action.

9. The winding machine according to claim 1, characterized in that: The multi-axis driving module, the tension dynamic compensation module, the intelligent detection module and the yarn breakage self-repairing module are connected to the control system via the EtherCAT communication protocol.

10. The winding machine according to claim 1, characterized in that: The control system also includes a human-machine interface for parameter setting, status monitoring, and data docking with the MES system.

Images