A needle wear testing device and method

Abstract

The present application relates to a kind of knitting needle wear test device and method, including test platform, line barrel, yarn accumulator, tension sensor, knitting needle fixture, microscopic magnifying glass, backend equipment and yarn winding mechanism, convolutional neural network model and data processing software are installed in backend equipment, the image photographed by microscopic magnifying glass is learned in depth by convolutional neural network model, to automatically judge whether knitting needle is worn or broken according to the image photographed by microscopic magnifying glass, while the fly of yarn after needle hook wear can be observed in real time, the influence of fly of yarn on fabric quality is judged by deep learning, whether knitting needle is failure is predicted, provide basis for the service life of knitting needle.

Description

Technical Field

[0001] This invention relates to the field of knitting needle wear testing technology, specifically a testing device and method for knitting needle hook wear failure and predicting service life. Background Technology

[0002] Currently, the textile industry still relies on traditional manual methods to determine the failure of knitting needles: judging based on the surface morphology of finished fabrics, or stopping the machine to disassemble and inspect the working condition of the knitting needles on-site. This type of detection mode is not only low in automation and highly subjective, but also causes production interruptions, seriously restricting the overall production efficiency of the machine.

[0003] Existing patented technologies cannot achieve online real-time detection of the degree of needle hook wear and its impact on yarn feathers, cannot provide prediction of needle lifespan, and are difficult to achieve periodic prediction and timely replacement of needles and avoid production capacity loss caused by downtime inspection. At the same time, there is no quantitative rating of the comprehensive performance of multiple types of needles, and it cannot provide data support for needle selection and operation and maintenance management. Summary of the Invention

[0004] This invention addresses the shortcomings of existing technologies by providing a device and method for testing knitting needle wear. This method enhances the relative motion and contact conditions between yarn and knitting needles during actual circular knitting machine production. It detects the wear degree, wear pattern, and wear rate of key wear parts of the needle (especially the needle hook) under different parameter conditions. Based on the variation law of the friction coefficient, the wear condition of the needle hook, and the yarn feathering, the lifespan of the knitting needle is determined. The experimentally tested lifespan of the knitting needle is compared with its actual lifespan, and the relationship between experimental parameters such as yarn type and yarn tension and the experimental and actual lifespan of the knitting needle is found. This allows for the detection of the quality grade of the knitting needle and the prediction of its actual lifespan.

[0005] To achieve the above objectives, the present invention adopts the following technical solution: A knitting needle wear testing device includes a testing platform, a yarn delivery spool, a meter counter, a yarn storage device, a tension sensor, a knitting needle clamp, a microscope, a rear-end device, and a yarn take-up mechanism. The yarn delivery spool, meter counter, yarn storage device, tension sensor, knitting needle clamp, and microscope are all mounted on the testing platform. The meter counter is used to record the yarn travel. The knitting needle clamp holds the knitting needle. Tension sensors are provided on both the yarn delivery side and the yarn take-up side of the knitting needle. A tension display is provided on the testing platform to display the values ​​of the tension sensors on the yarn delivery side and the yarn take-up side. The yarn winding mechanism includes a reciprocating slide rail, a take-up drum, and a three-section shaft. The reciprocating slide rail is used to drive the three-section shaft to reciprocate, ensuring that the tension fluctuation is small during the yarn spiral winding process and improving the stability and reliability of the equipment signal output and acquisition. The three-section shaft is used to drive the take-up drum to rotate and wind up the yarn. The yarn sequentially passes through the lead-out spool, meter counter, yarn storer, lead-out side tension sensor, knitting needle, and take-up side tension sensor before winding onto the take-up spool. The knitting needle clamp can be adjusted in position. The yarn storer is used to adjust the yarn tension during needle wear testing using electromagnetic induction. The meter counter records the yarn travel. The lead-out side tension sensor and the take-up side tension sensor are used to detect the yarn tension on the lead-out and take-up sides of the needle respectively during needle wear testing and send the data to the back-end equipment. The microscope is used to acquire real-time images of needle wear and needle movement. Images of yarn feathers after hook wear are sent to a backend device, which displays the observation image under a microscope in real time. The backend device is equipped with a hook life prediction module and a real-time friction coefficient calculation module. The real-time friction coefficient calculation module calculates the real-time friction coefficient between the yarn and the knitting needle based on the real-time tension transmitted by the tension sensor on the yarn feed side and the tension sensor on the yarn take-up side. The knitting needle life prediction module judges the wear condition of the knitting needle and predicts the life of the knitting needle based on the real-time friction coefficient, images of the degree of knitting needle wear, and images of yarn feathers after hook wear. The needle life prediction module is a deep learning model trained using a training set. The training samples in the training set are images of needle wear with known needle wear and needle life, and images of yarn feathers after needle hook wear.

[0006] Furthermore, the test platform is provided with a mounting base, and the knitting needle clamp is fixed on the mounting base. The mounting base is used to drive the knitting needle clamp to move laterally and longitudinally.

[0007] Furthermore, the three-section shaft includes an upper shaft section, a middle shaft section, and a lower shaft section. The upper shaft section and the lower shaft section are rotatably connected to the mounting bracket on the reciprocating slide rail, respectively. A take-up drum is provided on the middle shaft section of the three-section shaft, and the lower shaft section of the three-section shaft is fixedly connected to the output end of the drive motor.

[0008] Furthermore, the needle clamp includes a clamp base, a main clamp plate, and a secondary clamp plate. The main clamp plate and the secondary clamp plate are used to clamp the needles to form a clamping body. The main clamp plate and the secondary clamp plate are provided with universal fixing grooves for fixing common types of needles. The clamp base is provided with a fixing cavity for fixing the clamping body. The clamping body is embedded in the fixing cavity.

[0009] A method for testing knitting needle wear, characterized by comprising the following steps: Step 1: The yarn is sequentially passed from the output spool through the meter wheel, yarn storage device, output side tension sensor, knitting needle, and take-up side tension sensor before being wound onto the take-up spool. The rotation speed of the take-up spool is set. While the take-up spool is rotating, the meter wheel calculates the yarn travel speed and yarn stroke. The initial yarn tension value is set. The electromagnetic induction intensity of the yarn storage device is adjusted to ensure a stable output of the yarn tension value. The output side tension sensor detects the yarn tension F1 before the yarn rubs against the knitting needle. The take-up side tension sensor detects the real-time yarn tension F2 after the yarn rubs against the knitting needle. The tension display shows the real-time values ​​of F1 and F2 and transmits the values ​​to the back-end equipment. Step 2: The protractor measures the wrap angle θ between the yarn and the needle at the initial state, on both the exit and entry sides. The real-time friction coefficient calculation module calculates the real-time friction force f between the yarn and the needle based on the filtered data, the tensions F1 and F2, and the wrap angle θ. When the exit and entry angles of the yarn are the same, the friction force calculation formula is as follows: ; f is the real-time frictional force between the yarn and the knitting needle; θ is the wrap angle between the contact point of the yarn with the needle on the lead-out side and the contact point on the take-up side; F1 represents the real-time tension of the yarn on the needle exit side. F2 is the real-time tension of the yarn on the take-off side of the knitting needle; Step 3: The real-time friction coefficient calculation module calculates the real-time friction coefficient μ between the yarn and the knitting needle based on the frictional force f between the yarn and the knitting needle, and plots a dotted line graph based on the real-time friction coefficient μ at continuous time points. Ignoring the weight of the knitting needle, the calculation formula is as follows: ; μ is the real-time friction coefficient between the yarn and the knitting needle; Step 4: The needle life prediction module determines the fatigue life of the needle based on the dot-line graph of the real-time friction coefficient μ, the images of needle wear taken with a microscope, and the images of yarn feathers after the needle hook wears. Step 5: Select one of the following factors to change: yarn material, wear time, yarn travel speed, initial yarn tension, needle material, and the wrap angle θ between the yarn and the needle. Change each factor at least once. Go back to step 1 and repeat steps 1 to 4. The wrap angle θ between the yarn and the needle is changed by changing the installation position of the needle clamp. Step 6: Compare the friction coefficient μ line graphs for the same variable factor one by one, and determine the influence of different experimental parameters on the fatigue life of the knitting needle.

[0010] The beneficial effects of this invention are as follows: 1. This invention focuses on the wear detection of key wear parts of knitting needles (especially the needle hook), achieving precise testing through specialized experimental instruments. The tension yarn feeder uses PID control for electromagnetic force to apply stable tension to the yarn. Dual tension sensors positioned before and after the knitting needle accurately measure the tension difference. Combined with innovative calculation methods, the real-time friction coefficient of the knitting needle can be scientifically and quantitatively analyzed to determine the wear condition of the needle hook, overcoming the limitations of traditional devices that only provide qualitative observation. The knitting needle clamp adopts a universal structural design, adaptable to various models and specifications of knitting needles, offering wide applicability and strong compatibility. Simultaneously, the knitting needle fixing table allows for longitudinal and lateral adjustment, precisely controlling the yarn wrap angle parameters to simulate the spatial positional relationship between the knitting needle and yarn on a circular knitting machine. This highly replicates the actual maximum force state of the knitting needle on the machine, ensuring the authenticity and reliability of the test results. This invention can calculate the real-time change in the friction coefficient between the yarn and the needle hook, and determine the degree of needle hook wear and predict its service life by observing changes in the surface roughness of the needle hook. A high-magnification microscope is mounted on the side of the needle holder, with the lens directly facing the working area of ​​the needle. This allows for real-time microscopic observation of the wear morphology, degree of wear, and deformation state of the needle hook and working surface throughout the entire process. It clearly captures the entire wear evolution process, providing intuitive and visual evidence for needle wear mechanism analysis, operational condition monitoring, and lifespan assessment. Simultaneously, it can observe the yarn feathering after needle hook wear in real time. Through deep learning, it detects the yarn feathering and the impact of yarn wear on the fabric, determining the effect of needle hook wear on yarn feathering and the resulting fabric defects. This allows for the determination of the needle's lifespan, providing a basis for needle lifespan prediction.

[0011] 2. This invention has a simple structure and is easy to operate. The loading and unloading process of knitting needles and yarn is efficient. The application of a visual online detection method on the same equipment to determine the wear degree and service life of the needle hook not only provides a scientific tool for knitting needle quality assessment, service life judgment and design improvement, but also provides reliable data support for the study of wear mechanism in related fields.

[0012] 3. This invention allows for flexible adjustment of parameters such as yarn type, needle type, yarn tension, yarn-needle hook wrap angle, and yarn movement speed. This not only precisely matches different actual production conditions but also accelerates the experimental process through parameter optimization, significantly improving testing efficiency. Based on its core function of detecting needle hook wear, this instrument can also be extended to wear tests on various types of yarns and other hard, wear-resistant objects, demonstrating versatility and application flexibility.

[0013] The present invention will now be described in detail with reference to the accompanying drawings and embodiments. Attached Figure Description

[0014] Figure 1 This is a front view schematic diagram of the present invention; Figure 2A three-dimensional structural diagram of a knitting needle clamp; Figure 3 This is a schematic diagram of a three-dimensional structure with three shaft segments; Figure 4 The graph shows the real-time friction coefficient μ (red indicates the filtered graph). Figure 5 A schematic diagram of the knitting needle structure after processing by a convolutional neural network model; Figure 6 A schematic diagram showing the wear level of knitting needles as they approach their limit of fatigue life; Figure 7 This is a diagram showing the yarn feather structure on the needle hook's outgoing and take-up sides (left side is the take-up side, right side is the outgoing side).

[0015] The components represented by the numbers in the attached diagram are as follows: 1. Testing platform; 2. Yarn delivery drum; 3. Yarn storage device; 4. Knitting needle clamp; 401. Clamp base; 402. Main clamp plate; 403. Secondary clamp plate; 5. Microscope; 6. Back-end equipment; 7. Yarn winding mechanism; 701. Reciprocating slide rail; 702. Take-up drum; 703. Three-section shaft; 8. Yarn delivery side tension sensor; 9. Take-up side tension sensor; 10. Tension display instrument; 11. Mounting base; 12. Meter counting wheel. Detailed Implementation

[0016] The principles and features of the present invention are described below with reference to the accompanying drawings. The examples given are only for explaining the present invention and are not intended to limit the scope of the present invention.

[0017] like Figure 1-3 As shown, a knitting needle wear testing device includes a testing platform 1, a yarn feed spool 2, a meter counter 12, a yarn storage device 3, a tension sensor, a knitting needle clamp 4, a microscope 5, a rear-end device 6, and a yarn winding mechanism 7. The yarn feed spool 2, the meter counter 12, the yarn storage device 3, the tension sensor, the knitting needle clamp 4, and the microscope 5 are all mounted on the testing platform 1. The knitting needle clamp 4 holds a knitting needle. Tension sensors are provided on both the yarn feed side and the yarn winding side of the knitting needle. A tension display 10 is provided on the testing platform 1. The tension display 10 is used to display the values ​​of the yarn feed side tension sensor 8 and the yarn winding side tension sensor 9.

[0018] The yarn winding mechanism 7 includes a reciprocating slide rail 701, a take-up drum 702, and a three-section shaft 703. The reciprocating slide rail 701 is used to drive the three-section shaft 703 to reciprocate, and the three-section shaft 703 is used to drive the take-up drum 702 to rotate and wind up the yarn. The yarn sequentially passes through the lead-out spool 2, the meter counter 12, the yarn storage device 3, the lead-out side tension sensor 8, the knitting needle, and the take-up side tension sensor 9 before winding onto the take-up spool 702. The knitting needle clamp 4 can be adjusted in position. The meter counter 12 records the yarn travel. The yarn storage device 3 is used to adjust the yarn tension during the knitting needle wear test using the principle of electromagnetic induction. The lead-out side tension sensor 8 and the take-up side tension sensor 9 are used to detect the yarn tension on the lead-out and take-up sides of the knitting needle respectively during the knitting needle wear test and send the data to the back-end device 6. The microscope 5 is used to acquire images of knitting needle wear and yarn feathers after needle hook wear in real time and send them to the back-end device 6. The back-end device 6 is used to display the images in real time. The observation image from the micro-magnifying glass 5 shows that the back-end device 6 is equipped with a hook needle life prediction module and a real-time friction coefficient calculation module. The real-time friction coefficient calculation module is used to calculate the real-time friction coefficient between the yarn and the knitting needle based on the real-time tension transmitted by the yarn-leading tension sensor 8 and the yarn-take-up tension sensor 9. The knitting needle life prediction module is used to determine the wear condition of the knitting needle and predict its lifespan based on the real-time friction coefficient, images of the knitting needle wear degree, and images of yarn feathers after the hook wear. The knitting needle life prediction module is a deep learning model obtained by training a training set. The training samples in the training set are images of knitting needle wear with known wear conditions and lifespan, and images of yarn feathers after the hook wear.

[0019] Specifically, the needle life prediction module uses a convolutional neural network model to perform deep learning on the images captured by the microscope 5. Based on the images captured by the microscope 5, it automatically determines the wear degree of the needle and the yarn feathering after the needle hook is worn. Through deep learning, it determines the relationship between the yarn feathering and the lifespan of the needle, providing a basis for predicting the lifespan of the needle. The real-time friction coefficient calculation module is a known data processing software. It performs real-time conversion and calculation of the tension F1 on the outgoing side and the tension F2 on the receiving side to obtain the real-time friction force f and the real-time friction coefficient μ. It also automatically plots a dot-line graph of the relationship between the real-time friction coefficient μ and time, with time as the horizontal axis and the real-time friction coefficient as the vertical axis.

[0020] The test platform 1 is provided with a mounting base 11, and the knitting needle clamp 4 is fixed on the mounting base 11. The knitting needle clamp 4 can move laterally and longitudinally on the mounting base 11.

[0021] Specifically, the mounting base 11 is provided with a longitudinal slide rail and a transverse slide rail. The transverse slide rail is installed on the slider of the longitudinal slide rail, and the knitting needle clamp 4 is installed on the slider of the transverse slide rail.

[0022] In one embodiment, the three-section shaft 703 includes an upper shaft section, a middle shaft section, and a lower shaft section. The upper shaft section and the lower shaft section are rotatably connected to the mounting bracket on the reciprocating slide rail 701, respectively. A take-up drum 702 is provided on the middle shaft section of the three-section shaft 703, and the lower shaft section of the three-section shaft 703 is fixedly connected to the output end of the drive motor. The three-section shaft design can realize quick replacement of the bobbin and avoid the disassembly process of the motor.

[0023] In one embodiment, the knitting needle clamp 4 includes a clamp base 401, a main clamping plate 402, and a secondary clamping plate 403. The main clamping plate 402 and the secondary clamping plate 403 are used to clamp knitting needles to form a clamping body. The main clamping plate 402 and the secondary clamping plate 403 are provided with universal fixing grooves for fixing common types of knitting needles. The clamp base 401 is provided with a fixing cavity for fixing the clamping body. The clamping body is embedded in the fixing cavity.

[0024] A method for testing knitting needle wear, characterized by comprising the following steps: Step 1: The yarn is sequentially passed from the output spool 2 through the meter wheel 12, the yarn storage device 3, the output side tension sensor 8, the knitting needle, and the take-up side tension sensor 9 before being wound onto the take-up spool 702. The rotation speed of the take-up spool is set. When the take-up spool rotates, the meter wheel 12 calculates the yarn travel distance and yarn travel speed. The initial yarn tension value is set. The electromagnetic induction intensity of the yarn storage device is adjusted to ensure a stable output of the yarn tension value. The output side tension sensor 8 detects the yarn tension F1 before the yarn rubs against the knitting needle, and the take-up side tension sensor 9 detects the yarn tension F2 after the yarn rubs against the knitting needle. The tension display shows the real-time values ​​of F1 and F2 and transmits the values ​​to the back-end equipment. Step 2: Use a protractor to measure the wrap angle θ between the yarn and the needle at the initial state, on the exit side and the entry side. Subsequent minor changes in the wrap angle θ are ignored. The data processing software calculates the real-time friction force f between the yarn and the needle based on the filtered real-time tension F1 and tension F2 data and the wrap angle θ. When the angles on the exit side and the entry side of the yarn are the same, the friction force calculation formula (1) is as follows: (1) Step 3: The data processing software calculates the real-time friction coefficient μ between the yarn and the knitting needle based on the friction force f between the yarn and the knitting needle. Ignoring the weight of the knitting needle, the calculation formula (2) is as follows: (2) And plot a dotted line graph based on the real-time friction coefficient μ at continuous time intervals; Step 4: The convolutional neural network model determines the fatigue life of the knitting needle based on the dot-line graph of the real-time friction coefficient μ, the images of needle wear taken with a microscope, and the images of yarn feathers after the needle hook wears. Step 5: Select one of the following factors to change: yarn material, wear time, yarn travel speed, yarn initial tension, needle material, and the wrap angle θ between the yarn and the needle. Change each factor at least once. Go back to step 1 and repeat steps 1 to 4. The wrap angle θ between the yarn and the needle is changed by changing the installation position of the needle clamp 4. Step 6: Compare the friction coefficient μ line graphs for the same variable factor one by one, and determine the influence of different experimental parameters on the fatigue life of the knitting needle.

[0025] like Figure 4 As shown, adjusting the yarn feeder to control the yarn tension at 1N, the real-time friction coefficient μ fluctuates significantly in the first 1-2 hours, which is the break-in period for the knitting needle hooks. After 2-3 hours, the surface roughness of the needle hooks decreases, gradually becoming smoother, and the real-time friction coefficient μ gradually stabilizes. The rapid fluctuation period from 3-3.5 hours marks the beginning of deep wear on the needle hooks, with increased fluctuations in the real-time friction coefficient, followed by a gradual stabilization period. Figure 5 and Figure 6 As shown, after 5.5 hours, the abrasion marks deepened and widened, and the coefficient of friction fluctuated significantly. At this point, yarn feathering became severe. Figure 7 As shown, the yarn produces a large number of feathers after passing through the knitting needles. For high-quality fabrics, the knitting needles have become obsolete and can be discarded. In actual production, the actual fatigue life of the knitting needles can be determined by comparing the real-time yarn tension detection.

[0026] The workflow of this invention: Step 1: The yarn is sequentially passed from the output spool 2 through the meter wheel 12, the yarn storage device 3, the output side tension sensor 8, the knitting needle, and the take-up side tension sensor 9 before being wound onto the take-up spool 702. The rotation speed of the take-up spool is set. When the take-up spool rotates, the meter wheel 12 calculates the yarn travel speed and yarn stroke, and the initial yarn tension value is set. The electromagnetic induction intensity of the yarn storage device is adjusted to ensure a stable output of the yarn tension value. The output side tension sensor 8 detects the yarn tension F1 before the yarn rubs against the knitting needle, and the take-up side tension sensor 9 detects the yarn tension F2 after the yarn rubs against the knitting needle. The tension display shows the real-time values ​​of F1 and F2 and transmits the values ​​to the back-end equipment. Step 2: Use a protractor to measure the wrap angle θ between the yarn and the needle at the initial state, on the side where the yarn exits and the side where the yarn enters. The data processing software calculates the real-time frictional force f between the yarn and the needle based on the filtered real-time tension F1 and tension F2 data and the wrap angle θ. Step 3: The data processing software calculates the real-time friction coefficient μ between the yarn and the knitting needle based on the friction force f between the yarn and the knitting needle, and plots a dot-line graph based on the real-time friction coefficient μ at continuous time intervals. Step 4: The convolutional neural network model predicts the fatigue life of the knitting needle based on the dot-line graph of the real-time friction coefficient μ, the real-time monitoring image of the knitting needle under the microscope 5, and the yarn feathering change analysis based on deep learning. Step 5: Select one of the following factors to change: yarn material, wear time, yarn travel speed, yarn initial tension, needle material, and the wrap angle θ between the yarn and the needle. Change each factor at least once. Go back to step 1 and repeat steps 1 to 4. The wrap angle θ between the yarn and the needle is changed by changing the installation position of the needle clamp 4. Step 6: Compare the friction coefficient μ line graphs of the same variable factor one by one, and determine the influence of different yarn materials and test parameters such as the wrap angle θ between the yarn and the knitting needle on the yarn feathering, needle hook wear, etc. within the same time period. Obtain the influence law of each parameter on the fatigue life of the knitting needle, and establish a knitting needle fatigue life prediction model.

[0027] The above description provides examples of the preferred embodiments of the present invention. Parts not detailed herein are common knowledge to those skilled in the art. The scope of protection of the present invention is determined by the claims. Any equivalent modifications based on the technical teachings of the present invention are also within the scope of protection of the present invention.

Claims

1. A device for testing the wear of knitting needles, characterized in that, The test platform (1), yarn spool (2), yarn storage device (3), tension sensor, needle clamp (4), microscope (5), back-end equipment (6) and yarn winding mechanism (7) are all installed on the test platform (1). The needle clamp (4) holds the knitting needle. The yarn spool (3), tension sensor, needle clamp (4), microscope (5) and back-end equipment (6) are all installed on the test platform (1). The needle clamp (4) holds the knitting needle. The yarn spool tension sensor (8) and the yarn winding tension sensor (9) are respectively provided on the yarn spool side and the yarn winding side of the knitting needle. The test platform (1) is equipped with a tension display device (10), which is used to display the values ​​of the yarn spool tension sensor (8) and the yarn winding tension sensor (9). The yarn winding mechanism (7) includes a reciprocating slide rail (701), a take-up drum (702), and a three-section shaft (703). The reciprocating slide rail (701) is used to drive the three-section shaft (703) to reciprocate, and the three-section shaft (703) is used to drive the take-up drum (702) to rotate and wind up the yarn. The yarn passes sequentially through the lead-out spool (2), the meter wheel (12), the yarn storage device (3), the lead-out side tension sensor (8), the knitting needle, and the take-up side tension sensor (9) before being wound onto the take-up spool (702). The knitting needle clamp (4) can be adjusted in its installation position. The yarn storage device (3) is used to adjust the yarn tension during the knitting needle wear test by means of electromagnetic induction. The lead-out side tension sensor (8) and the take-up side tension sensor (9) are used to detect the yarn tension on the lead-out side and the take-up side of the knitting needle respectively during the knitting needle wear test and send it to the back-end device (6). The microscopic magnifying glass (5) is used to collect the knitting needle wear diagram in real time. Images of yarn feathers after wear of the needle and hook are sent to the back-end device (6). The back-end device (6) is used to display the observation screen of the microscope (5) in real time. The back-end device (6) is equipped with a hook life prediction module and a real-time friction coefficient calculation module. The real-time friction coefficient calculation module is used to calculate the real-time friction coefficient between the yarn and the knitting needle based on the real-time tension transmitted by the tension sensor (8) on the yarn exit side and the tension sensor (9) on the yarn take-up side. The knitting needle life prediction module is used to judge the wear of the knitting needle and predict the life of the knitting needle based on the real-time friction coefficient, the knitting needle wear degree image and the yarn feathers after the needle hook wear. The needle life prediction module is a deep learning model trained using a training set. The training samples in the training set are images of needle wear with known needle wear and needle life, and images of yarn feathers after needle hook wear.

2. The needle wear testing device according to claim 1, characterized in that, The test platform (1) is provided with a sliding mounting base (11), and the knitting needle clamp (4) is fixed on the mounting base (11). The mounting base (11) is used to drive the knitting needle clamp (4) to move laterally and longitudinally.

3. The needle wear testing device according to claim 1, characterized in that, The three-section shaft (703) includes an upper shaft section, a middle shaft section and a lower shaft section. The upper shaft section and the lower shaft section are rotatably connected to the mounting bracket on the reciprocating slide rail (701). A take-up drum (702) is provided on the middle shaft section of the three-section shaft (703). The lower shaft section of the three-section shaft (703) is fixedly connected to the output end of the drive motor.

4. The needle wear testing device according to claim 1, characterized in that, The needle clamp (4) includes a clamp base (401), a main clamp plate (402) and a secondary clamp plate (403). The main clamp plate (402) and the secondary clamp plate (403) are used to clamp the needles to form a clamping body. The main clamp plate (402) and the secondary clamp plate (403) are provided with universal fixing grooves for fixing common types of needles. The clamp base (401) is provided with a fixing cavity for fixing the clamping body. The clamping body is embedded in the fixing cavity.

5. A method for testing the abrasion resistance of knitting needles and yarns, characterized in that, Includes the following steps: Step 1: The yarn is sequentially passed from the output spool through the meter wheel, yarn storage device, output side tension sensor, knitting needle, and take-up side tension sensor before being wound onto the take-up spool. The rotation speed of the take-up spool is set. While the take-up spool is rotating, the meter wheel calculates the yarn travel speed and yarn stroke. The initial yarn tension value is set. The electromagnetic induction intensity of the yarn storage device is adjusted to ensure a stable output of the yarn tension value. The output side tension sensor detects the yarn tension F1 before the yarn rubs against the knitting needle, and the take-up side tension sensor detects the yarn tension F2 after the yarn rubs against the knitting needle. The tension display shows the real-time values ​​of F1 and F2 and transmits the values ​​to the back-end equipment. Step 2: The protractor measures the wrap angle θ between the yarn and the needle at the initial state, on both the exit and entry sides. The real-time friction coefficient calculation module calculates the real-time friction force f between the yarn and the needle based on the processed real-time tensions F1 and F2 and the wrap angle θ. At this point, it is assumed that the angles on the exit and entry sides of the yarn are symmetrical. The calculation formula is as follows: ； f is the real-time frictional force between the yarn and the knitting needle; θ is the wrap angle between the contact point of the yarn with the needle on the lead-out side and the contact point on the take-up side; F1 represents the real-time tension of the yarn on the needle exit side; F2 is the real-time tension of the yarn on the take-off side of the knitting needle; Step 3: The real-time friction coefficient calculation module calculates the real-time friction coefficient μ between the yarn and the knitting needle based on the frictional force f between the yarn and the knitting needle, and plots a dotted line graph based on the real-time friction coefficient μ at continuous time points. Ignoring the weight of the knitting needle, the calculation formula is as follows: ； μ is the real-time friction coefficient between the yarn and the knitting needle; Step 4: The needle life prediction module determines the fatigue life of the needle based on the dot-line graph of the real-time friction coefficient μ, the images of needle wear taken with a microscope, and the images of yarn feathers after the needle hook wears. Step 5: Select one of the following factors to change: yarn material, wear time, yarn travel speed, initial yarn tension, needle material, and the wrap angle θ between the yarn and the needle. Change each factor at least once. Go back to step 1 and repeat steps 1 to 4. The wrap angle θ between the yarn and the needle is changed by changing the installation position of the needle clamp. Step 6: Compare the friction coefficient μ line graphs for the same variable factor one by one, and determine the influence of different experimental parameters on the fatigue life of the knitting needle.

Images