CN-121994809-A - On-line detection method, equipment and system for polyethylene fiber yarns

CN121994809ACN 121994809 ACN121994809 ACN 121994809ACN-121994809-A

Abstract

The invention relates to an online detection method, equipment and a system for polyethylene fiber yarns, and belongs to the technical field of intelligent manufacturing of high-performance fibers. The device comprises a yarn color detection module, a surface defect detection module, an Optical Coherence Tomography (OCT) module and an ultrasonic diameter detection module, and can carry out multidimensional detection on the appearance and internal structure defects of the yarn in the production process. By constructing a process parameter optimization model based on flaw indexes and diameter errors, the system can dynamically adjust key process parameters such as melting temperature, high shear rate, cooling rate and the like according to real-time detection data, and closed-loop control is realized. The process parameter updating adopts the joint optimization of a sensitivity matrix and a weight matrix, and is provided with a parameter constraint mechanism for preventing boundary crossing and system oscillation. The whole set of system can promote the quality uniformity and the manufacturing stability of polyethylene fiber yarns, and is suitable for continuous and intelligent production scenes.

Inventors

Di Junxian
ZHOU XINJI
ZHU JIANJUN
NIU YANFENG
HUANG HUIJUN
LIU PENG

Assignees

江苏九州星际新材料有限公司
浙江理工大学

Dates

Publication Date: 20260508
Application Date: 20250723

Claims (7)

1. The on-line detection equipment for the polyethylene fibers is characterized by comprising an upper box body, (2) a first guide wheel, (12) a second guide wheel, and an image detection module, wherein the upper box body is fixed on the upper box body, the image detection module comprises (14) an industrial camera which is fixed in a closed space of the upper box body, an LED strip-shaped light source is arranged in the upper box body, and the LED strip-shaped light source can provide proper light source conditions, and a code sprayer (15). (11) The upper box body and the lower box body can be separated by the partition board, the support board (15) is arranged in the lower box body, the tension wheel (5) and the first reversing wheel (3) are arranged on the support board, (8) the second reversing wheel, (10) the third guiding wheel, (9) the Optical Coherence Tomography (OCT) module, (7) the fourth guiding wheel, (6) the ultrasonic diameter detection module and (17) the industrial camera are also arranged, and the ultrasonic diameter detection device further comprises a data processing module (16) for receiving defect data transmitted by the fiber surface defect detection module, the Optical Coherence Tomography (OCT) module and the ultrasonic diameter detection module, calculating the generation rate of defects at intervals, and adjusting the technological parameters to reduce the generation of the defects if the defect rate exceeds a threshold value.
2. The on-line detection device for the polyethylene fibers according to claim 1, wherein the on-line detection device comprises the following steps that a light-adjustable module adjusts the wavelength of a strip-shaped LED light source according to the color of the polyethylene fibers through an industrial camera, a defect detection module detects images, if the light-adjustable module is not qualified, information is fed back to the data processor, the data processing module performs statistical analysis on data, a code sprayer is controlled to mark defective parts, if the light-adjustable module is qualified, the operation is continued, the defect position of the fiber is judged through an Optical Coherence Tomography (OCT) module, if the light-adjustable module is qualified, the defect position is marked through the code sprayer, whether the diameter meets the standard or not is judged through an ultrasonic diameter detection module, if the light-adjustable module is qualified, the operation is continued, and if the light-adjustable module is not qualified, the light-adjustable light-emitting diode is marked by the defect position.
3. The method for on-line detection of polyethylene fibers according to claim 2, wherein the adjustable light module adjusts the light source, comprising the steps of: The industrial camera (17) sends yarn color data to the data processing module (16), and the data processing module (16) adjusts the wavelength of the strip-shaped LED light source (13) in the adjustable light module to create illumination conditions most suitable for the defect detection module.
4. The method for on-line detection of polyethylene fibers according to claim 2, wherein the defect detection module judges whether the collected image is acceptable based on the image data, comprising the steps of: acquiring an image before defect detection is carried out on the polyethylene fiber yarn on-line detection equipment; Transmitting the image into a defect detection module, wherein the defect detection module detects the image by adopting a convolutional neural network; If the detection is qualified, continuing to operate; If the defect signal is unqualified, the defect signal is transmitted into a data processing module, the data processing module performs data statistics, and meanwhile, the code sprayer (15) is controlled to perform defect marking.
5. The method for on-line detection of polyethylene fibers according to claim 2, wherein the Optical Coherence Tomography (OCT) module performs the defect detection inside the fibers, comprising the steps of: performing low-resolution scanning based on polyethylene fibers to locate suspected defect areas; performing dense sampling in a defect area, and simultaneously performing multi-angle scanning and rotating fiber repeated scanning at 30 degrees and 60 degrees, so that the defect three-dimensional reconstruction accuracy is enhanced; the obtained data is fed back to a data processing module, and the data processing module processes the data; Performing Fourier transform on the acquired slice data to generate a 2D image; denoising the generated 2D image by adopting mean filtering, and enhancing the denoised image by histogram equalization; If the image is detected to be qualified, continuing to operate; if the data is not qualified, the data is fed back to the data processing module, and the code sprayer (15) is controlled to perform defect marking.
6. The on-line detection method of polyethylene fiber according to claim 2, wherein the ultrasonic diameter detection module detects the diameter of the polyethylene fiber, comprising the steps of: The ultrasonic probe transmits high-frequency ultrasonic wave beams, passes through the polyethylene fibers and reaches a receiving end, the ultrasonic waves are partially reflected at the interface of the fibers and the coupling medium and partially penetrate to the other side, a receiving probe at the other side captures transmission signals, and a transmitting probe can receive the reflected signals, and the method comprises the following steps: , In order to be the propagation velocity of the ultrasonic wave, And For the propagation time of the ultrasonic wave to the interface on both sides, For error compensation, if the diameter error rate is Exceeding the set qualification threshold value The error rate formula is as follows , Is the optimal diameter.
7. An on-line detection method of polyethylene fiber according to any one of claims 4 to 6, wherein the data processing module processes the data, comprising the steps of: If defects exist on the surface of the fiber, the error rate of the fiber diameter and the inside of the fiber, the defect information is fed back to the data processing module; the data processing module counts according to the average value of flaw indexes in a period of time, and completes optimization of technological parameters through a parameter optimization model; the flaw index can be obtained according to the technological parameters and the related fitting coefficients, and the flaw index is as follows: In the formula (I), in the formula (II), Is the first The process parameters of the method are that, As an optimal value for the parameter, , , For the fitting coefficients, determined by experimental data regression, Is a random error term; The process parameter optimization model is as follows: In the formula (I), in the formula (II), Is the technological parameter Is used for adjusting the adjustment amount of the (a), In order for the rate of learning to be high, Is the first The index of the defects is that, Is a flaw For parameters Is the sensitivity (partial derivative), In order to actually detect the flaw value, Is the target flaw value; When (when) Or (b) When in use; In order to ensure that the system operates stably, the constraint of the process parameters is carried out, and the constraint model is as follows: ; ; Is a self-adaptive adjustment factor, controls adjustment force and prevents oscillation; Is unconstrained process parameter adjustment; Is an adjusting coefficient for controlling the adjusting force, when the error is small, Allows larger adjustment, ensures rapid convergence, and when the error is larger, The adjustment amplitude is reduced, and instability is avoided.

Description

On-line detection method, equipment and system for polyethylene fiber yarns Technical Field The invention belongs to the technical field of yarn online detection, and particularly relates to a method, equipment and a system for online detection of polyethylene fiber yarns. Background With the wide application of high-performance fiber materials in the fields of aerospace, protective equipment, high-end manufacturing and the like, higher requirements are put on the quality stability and consistency of fiber products. Polyethylene fibers (e.g., ultra high molecular weight polyethylene fibers, UHMWPE) are important in industrial applications as an engineering fiber having excellent properties such as high strength, high modulus, corrosion resistance, etc. The production process of polyethylene fiber at present generally comprises a plurality of process links such as melt spinning, stretching shaping, cooling shaping and the like. In the actual production process, the problems of surface flaws, bubbles, uneven thickness and the like of the fiber are easily caused due to the fluctuation of process parameters such as melting temperature, drawing multiplying power, cooling speed and the like, thereby influencing the mechanical property and the service life of the fiber. The traditional quality control means depend on manual sampling detection or single on-line sensing technology, cannot realize full-process, full-size and real-time defect detection on fibers, lacks an effective process parameter feedback adjustment mechanism, and is difficult to meet the requirements of intelligent manufacturing. Therefore, there is a need for a polyethylene fiber online detection system that combines a multi-mode detection technology and a self-adaptive optimization algorithm, and is capable of comprehensively detecting color, surface defects, internal structure and diameter changes of the fiber, dynamically optimizing a production process based on detection results, and realizing closed-loop control and fine management, thereby improving consistency and process stability of fiber products. Disclosure of Invention The invention designs an online detection method, equipment and a system for polyethylene fiber yarns, which can effectively ensure the quality of the polyethylene fiber yarns in the production and manufacturing process. In order to achieve the above purpose, the invention is realized by the following technical scheme: The polyethylene fiber yarn on-line detection equipment comprises an upper box body, wherein a yarn color detection module is arranged on a wire inlet of the upper box body; The upper box body is provided with a first guide wheel and a second guide wheel, a cavity is formed in the upper box body, a first wire guide hole and a second wire guide hole are formed in the upper box body, and a surface defect detection module which comprises an industrial camera, an LED strip-shaped light source and a code sprayer is arranged in the upper box body; The upper box body and the lower box body are separated by a baffle plate, and a third wire guide hole is formed in the baffle plate; A supporting plate is arranged in the lower box body, a third guide wheel, a fourth guide wheel, a first reversing wheel, a second reversing wheel, a tensioning wheel, an ultrasonic diameter detection module and an Optical Coherence Tomography (OCT) module are arranged on the supporting plate; the on-line detection method of the polyethylene fiber yarn is based on the equipment and comprises the following steps: judging the wavelength of the yarn color adjusting LED strip light source; passing through the first wire guide around the second guide wheel; the surface defect detection module detects; If the operation is qualified, continuing to operate; If the defect data are unqualified, feeding the defect data back to the data processing module; the data processing module matches corresponding technological parameters in a database according to the defect type; According to The model calculates the flaw index, wherein,Is the firstThe process parameters of the method are that,As an optimal value for the parameter,,,For the fitting coefficients, determined by experimental data regression,Is a random error term; Optimizing the data model according to the technological parameters In the formula,Is the technological parameterIs used for adjusting the adjustment amount of the (a),In order for the rate of learning to be high,Is the firstThe index of the defects is that,Is a flawFor parametersIs the sensitivity (partial derivative),In order to actually detect the flaw value,Is the target flaw value; If a plurality of flaws exist, according to Process parameter updating is carried out, whereinAdjusting vectors for process parameters,For the adjustment value of the melting temperature,For an adjustment value of the high shear rate,In order to achieve a cooling rate, Is the flaw deviation vector,In order to achieve the air bubble density,The target