Fiber fabric production quality monitoring method and system
Through multi-stage monitoring of the fiber fabric production process, spinning uniformity, stretching and shaping stability and fabric structural integrity index are obtained, which solves the shortcomings of real-time physical parameter monitoring and comprehensive quality evaluation in the existing technology, and achieves high efficiency quality control and product performance improvement.
Patent Information
- Application Number
- CN202411003594.1
- Authority / Receiving Office
- CN · China
- Patent Type
- Applications(China)
- Current Assignee / Owner
- Filing Date
- 2024-07-25
- Publication Date
- 2024-11-15
AI Technical Summary
The prior art lacks direct monitoring of real-time physical parameters and comprehensive evaluation of comprehensive quality characteristics in fiber fabric production, resulting in poor quality control effect in the production process.
A fiber fabric production quality monitoring method is adopted to obtain spinning uniformity index, stretching stability index and fabric structural integrity index by conducting multi-stage monitoring of melt spinning, stretching and shaping stages, and through the integration Analyze and generate comprehensive quality monitoring index. The method includes image processing, convolutional neural networks, rheometers, online tension testing equipment and sensor technology to monitor fiber diameter, surface defects, melt flow, spinning speed, fiber strength and fabric structure in real time.
It has achieved comprehensive quality control of the fiber fabric production process, reduced the output of unqualified products, ensured the uniformity and performance of fibers comply with standards, and improved production efficiency and product quality.
Smart Images

Figure CN118966879A_ABST
Abstract
Description
Technical Field
[0001] The invention relates to the technical field of clothing fabrics, and in particular to a method and system for monitoring the production quality of fiber fabrics. Background Art
[0002] Polyester fiber, mainly referring to polyethylene terephthalate (PET) fiber, is a synthetic fiber made from raw materials such as petroleum, coal, air and water. This fiber is extremely popular in the textile industry due to its excellent durability, wrinkle resistance and shape retention. Polyester fiber can be made into filaments through a melt spinning process and further processed into a variety of textiles. In recent years, polyester fiber has been mainly made from plastic bottles. This method can convert a large amount of plastic waste into useful products, reducing landfill and environmental pollution.
[0003] There are many reasons why polyester is used to make fabrics. First, polyester has high strength and stretch, which allows it to withstand daily wear and multiple washings without being damaged. Second, it has good shape memory, which reduces wrinkles in clothing and makes it easier to maintain. In addition, polyester has excellent chemical stability, is not easily corroded, and is relatively low in cost, making it an ideal choice for mass production of clothing and home textiles.
[0004] It is very important to monitor the production process of polyester fiber fabrics, because this can ensure that the quality of the fabrics meets the standard specifications, reduce production defects and improve production efficiency. The monitoring process can help to timely detect problems on the production line, such as insufficient fiber strength, unstable melt temperature or mechanical failure, so as to take measures to prevent defective products from entering the market. In addition, a good production monitoring system can optimize the use of resources, reduce waste and ensure the sustainability of production.
[0005] Traditional polyester fabric production monitoring methods mainly rely on simple mechanical testing, which includes using a tensile machine to measure the strength and elongation of the fiber, and using tools such as thickness gauges and density gauges to check the physical properties of the fabric. Although this method that relies on direct measurement and observation has been used for decades, it is usually inefficient and difficult to achieve high-precision and real-time production quality control.
[0006] There are many quality monitoring methods for fiber fabric production in the existing technology. For example, in 2022, scholar Jiang Chunli proposed a polyester fiber production process parameter optimization method based on industrial Internet of Things services, and deeply explored the intelligent monitoring and optimization of the entire process of polyester fiber production, including polymerization, melt delivery and spinning. First, through intelligent data processing and analysis, the study established a two-way optimization model that allows real-time adjustment of process parameters in the production process to optimize quality and efficiency. Secondly, the study uses a data-driven approach to optimize the specific process parameters of polyester fibers around different production batches and specifications to reduce energy consumption and adapt to the needs of new product development. Further, by designing a multi-objective optimization algorithm, the study improves the efficiency and cost-effectiveness of services in the Internet of Things environment. Finally, an evaluation mechanism based on biological intelligence was developed to more comprehensively evaluate production services, and through online process parameter optimization, rapid response to customer needs and deep integration of product information were achieved; although this method improves the level of intelligence and efficiency of production, this method mainly focuses on data processing and analysis, and lacks direct monitoring of real-time physical parameter changes in the production process, such as temperature, pressure, etc., which may lead to insufficient adaptability and responsiveness of the monitoring system to actual production conditions. A Chinese patent with patent application number CN202210583793.9 proposes a home textile fiber fabric production quality monitoring system, and its method steps include four main modules: First, the first data acquisition module divides the fabric into multiple sub-areas and obtains the fabric thickness of each sub-area. Subsequently, the preliminary screening module performs a first quality assessment based on these data to screen out better fabrics. Then, the second data acquisition module further measures the fabric density and transmittance of each sub-area in these better fabrics, and records the synchronization distance between each adjacent group of sub-areas. Finally, the quality screening module performs a linear fit on the synchronization distance, evaluates the final quality of the fabric based on the degree of fit, and determines whether it meets the quality standards; although this method can achieve regional quality assessment, it cannot fully capture the overall quality characteristics of the fabric, especially when facing complex textures and materials. In addition, the system relies on the measurement of fabric thickness and density, which may not be sufficient to fully evaluate the functional properties of the fabric, such as strength and elasticity, resulting in the practical evaluation of the final product may not be accurate or comprehensive enough. The above methods lack direct monitoring of real-time physical parameters and comprehensive evaluation of comprehensive quality characteristics in fiber fabric production quality monitoring, resulting in poor comprehensive quality monitoring of the fiber fabric production process.
[0007] Therefore, a fiber fabric production quality monitoring method and system are proposed. Summary of the invention
[0008] The object of the present invention is to provide a fiber fabric production quality monitoring method and system. First, a first production monitoring and a first state monitoring are performed on the melt spinning stage to obtain a first production quality value and a first state quality value, and a spinning uniformity index is obtained by comprehensively analyzing the first production quality value and the first state quality; then, a second production monitoring and a second state monitoring are performed on the stretching and shaping stage to obtain a second production quality value and a second state quality value, and a stretching and shaping stability index is obtained by comprehensively analyzing the second production quality value and the second state quality; further, a third production monitoring and a third state monitoring are performed on the spinning stage to obtain a third production quality value and a third state quality value, and a fabric structure integrity index is obtained by comprehensively analyzing the third production quality value and the third state quality; finally, a comprehensive quality monitoring index is obtained by comprehensively analyzing the spinning uniformity index, the stretching and shaping stability index and the fabric structure integrity index.
[0009] To achieve the above object, the present invention provides the following technical solutions:
[0010] A fiber fabric production quality monitoring method, comprising:
[0011] Performing a first quality monitoring on the melt spinning stage to obtain a spinning uniformity index; the first quality monitoring includes a first production monitoring and a first state monitoring; the first production monitoring includes: obtaining a first image of the fiber after melt spinning, and obtaining a first production quality value by preprocessing and surface analyzing the first image; the first state monitoring is used to monitor melt fluidity and spinning speed during the melt spinning process to obtain a first state quality value; the spinning uniformity index is obtained by comprehensively analyzing the first state quality value and the first production quality value;
[0012] Performing a second quality monitoring on the stretching and shaping stage to obtain a stretching and shaping stability index; the second quality monitoring includes a second production monitoring and a second state monitoring;
[0013] Performing a third quality monitoring on the textile stage to obtain a fabric structure integrity index; the third quality monitoring includes a third production monitoring and a third state monitoring;
[0014] A comprehensive quality monitoring index is obtained by comprehensively analyzing the spinning uniformity index, the stretching setting stability index and the fabric structure integrity index.
[0015] The preprocessing includes graying, denoising and edge detection of the first image;
[0016] The specific analysis process of the surface analysis includes:
[0017] Using the Hough transform algorithm to detect the fiber edge, obtain the fiber diameter; comparing the fiber diameter with the standard diameter, and calculating the difference;
[0018] Using a convolutional neural network to detect appearance defects on the fiber surface; the appearance defects include breakage, agglomeration and foreign matter;
[0019] The defect rate is calculated; the calculation formula of the defect rate is: Among them, S1 is the total area of the defect area, S 总 is the total fiber area;
[0020] The calculation formula of the first production quality value is:
[0021]
[0022] Among them, Q1 is the first production quality value, N1 is the number of monitoring samples of the first production monitoring, and D i is the diameter of the ith fiber, D std It is a standard diameter size;
[0023] The melt fluidity is obtained by a rheometer; the spinning speed is obtained by a speed sensor;
[0024] The calculation formula of the first state quality value is:
[0025]
[0026] Among them, P1 is the first state quality value, S avg ,S std They represent the average and standard melt fluidity, σ v is the standard deviation of spinning speed, V avg is the average spinning speed;
[0027] The formula for the spinning uniformity index is:
[0028] F1=β1·Q1+β2·P1;
[0029] Among them, F1 is the spinning uniformity index; β1, β2 are weight coefficients, and β1+β2=1.
[0030] The second production monitoring includes an online tensile testing device for monitoring the strength and elongation of the fiber during the stretching process to obtain a second production quality value; the second state monitoring includes a thermal imager for monitoring the temperature and time during the heat setting process to obtain a second state quality value; the tensile setting stability index is obtained by comprehensively analyzing the second state quality value and the second production quality value;
[0031] The calculation formula of the second production quality value is:
[0032]
[0033] Where, Q2 is the second production quality value, N2 is the number of monitoring samples for the second production monitoring, and C j ,E j are the fiber strength and fiber elongation of the jth sample, respectively, D std ,E std are standard fiber strength and standard fiber elongation, respectively;
[0034] The calculation formula of the second state quality value is:
[0035]
[0036] Among them, P2 is the second state quality value, T avg ,t std represent the average measured temperature and actual heat setting time, T std ,t std They are standard temperature and standard heat setting time respectively;
[0037] The formula for the tensile setting stability index is:
[0038] F2=β3·Q2+β4·P2;
[0039] Among them, F2 is the tensile setting stability index; β3, β4 are weight coefficients, and β3+β4=1.
[0040] The third production monitoring includes acquiring a second image of the fabric, and acquiring a third production quality value by performing defect analysis on the second image; the defect analysis includes analyzing whether the fabric has broken yarn and wrong weaving; the third state monitoring includes a sensor group for monitoring the yarn tension and weaving speed during the weaving process to acquire a third state quality value; the sensor group includes a tension sensor and a speed sensor; the fabric structural integrity index is acquired by comprehensively analyzing the third state quality value and the third production quality value;
[0041] The specific process of the defect analysis includes:
[0042] Performing a preprocessing operation on the second image to obtain a third image; the preprocessing operation includes graying, denoising and contrast enhancement;
[0043] Performing feature detection on the third image; the feature detection includes edge detection and texture analysis; the edge detection includes using a Canny edge detection algorithm to identify yarn edges and obtain edge features; the texture analysis includes analyzing a texture pattern of the fabric through a gray-level co-occurrence matrix and obtaining texture features;
[0044] The edge features and the texture features are analyzed using a CNN model to obtain specific locations and types of defects; the types include broken yarn and mis-weaving;
[0045] The calculation formula of the third production quality value is:
[0046]
[0047] Among them, δ break ,δ miss are the weights of yarn breakage and misweaving, respectively, severity,break ,λ severity,miss are the severity factors of yarn breakage and misweaving, A total is the total area analyzed, A break A is the total area of the detected broken yarn defect area, miss is the total area of the misweaving defect region detected;
[0048] The calculation formula of the third state quality value is:
[0049]
[0050] Among them, P3 is the third state quality value, L avg ,VS avg Represent the average values of tension and weaving speed, l n ,l vs represent the standard deviation of tension data and weaving speed data respectively;
[0051] The formula for the fabric structural integrity index is:
[0052] F3=β5·Q3+β6·P3;
[0053] Wherein, F3 is the fabric structural integrity index; β5, β6 are weight coefficients, and β5+β6=1.
[0054] The calculation formula of the comprehensive quality monitoring index is:
[0055] F al =α1·F1+α2·F2+α3·F3;
[0056] Among them, F al is a comprehensive quality monitoring index, F1, F2, F3 are spinning uniformity index, stretching setting stability index and fabric structure integrity index respectively, α1, α2, α3 are the weights of spinning uniformity index, stretching setting stability index and fabric structure integrity index respectively, and α1>α2>α3.
[0057] A fiber fabric production quality monitoring system, comprising:
[0058] A melt spinning analysis unit, including a first production monitoring unit and a first state monitoring unit, for performing a first quality monitoring on the melt spinning stage to obtain a spinning uniformity index;
[0059] The stretching and shaping analysis unit includes a second production monitoring and a second state monitoring, which is used to perform a second quality monitoring on the stretching and shaping stage and obtain a stretching and shaping stability index;
[0060] a textile analysis unit, including a third production monitoring and a third state monitoring, for performing a third quality monitoring of the textile stage and obtaining a fabric structural integrity index;
[0061] The comprehensive analysis unit is used to obtain a comprehensive quality monitoring index; the comprehensive quality monitoring index is obtained by comprehensively analyzing a spinning uniformity index, a stretching and shaping stability index, and a fabric structure integrity index.
[0062] The first production monitoring includes: obtaining a first image of the fiber after melt spinning, and obtaining a first production quality value by preprocessing and surface analyzing the first image; the first state monitoring is used to monitor the melt fluidity and spinning speed during the melt spinning process to obtain a first state quality value.
[0063] The preprocessing includes graying, denoising and edge detection of the first image;
[0064] The specific analysis process of the surface analysis includes:
[0065] Using the Hough transform algorithm to detect the fiber edge, obtain the fiber diameter; comparing the fiber diameter with the standard diameter, and calculating the difference;
[0066] Using a convolutional neural network to detect appearance defects on the fiber surface; the appearance defects include breakage, agglomeration and foreign matter;
[0067] The defect rate is calculated; the calculation formula of the defect rate is: Among them, S1 is the total area of the defect area, S 总 is the total fiber area;
[0068] The calculation formula of the first production quality value is:
[0069]
[0070] Among them, Q1 is the first production quality value, N1 is the number of monitoring samples of the first production monitoring, and D i is the diameter of the ith fiber, D std It is a standard diameter size;
[0071] The melt fluidity is obtained by a rheometer; the spinning speed is obtained by a speed sensor;
[0072] The calculation formula of the first state quality value is:
[0073]
[0074] Among them, P1 is the first state quality value, S avg ,S std They represent the average and standard melt fluidity, σ v is the standard deviation of spinning speed, V avg is the average spinning speed.
[0075] The formula for the spinning uniformity index is:
[0076] F1=β1·Q1+β2·P1;
[0077] Among them, F1 is the spinning uniformity index; β1, β2 are weight coefficients, and β1+β2=1.
[0078] The second production monitoring includes an online tensile testing device for monitoring the strength and elongation of the fiber during the stretching process to obtain a second production quality value; the second state monitoring includes a thermal imager for monitoring the temperature and time during the heat setting process to obtain a second state quality value; the stretching setting stability index is obtained by comprehensively analyzing the second state quality value and the second production quality value.
[0079] Compared with the prior art, the present invention has the following beneficial effects:
[0080] 1. The present invention implements the first quality monitoring in the melt spinning stage, and comprehensively analyzes the spinning uniformity index to make the quality monitoring of the spinning process more comprehensive. By pre-processing and surface analyzing the image of the fiber after melt spinning, this method can accurately monitor the fiber diameter and surface defects, and effectively monitor the quality stability of the spinning process by combining the real-time data of melt fluidity and spinning speed. This monitoring helps ensure the uniformity and quality of the fiber during the production process and reduces the output of defective products.
[0081] 2. The present invention obtains the stretching and shaping stability index by implementing the second quality monitoring in the stretching and shaping stage; the stretching and shaping stability index is a quantitative representation of the quality control of the entire stretching and shaping process, which is helpful for helping manufacturers monitor and maintain the stable operation of the production line. The strength and elongation of the fiber and the temperature and time during the heat setting process are monitored by online tensile testing equipment and thermal imagers to ensure the quality monitoring accuracy of the stretching and shaping process. Through real-time monitoring of these key parameters, the present invention can promptly detect any deviations in the production process and ensure that the performance and quality of the fiber meet the predetermined standards.
[0082] 3. The third quality monitoring implemented in the textile stage of the present invention obtains the fabric structural integrity index, ensuring the effective monitoring of the fabric structural integrity. By integrating defect analysis with status monitoring, this method can accurately identify structural defects in the fabric, including yarn breakage and misweaving, and monitor tension and weaving speed in real time, and ensure the high quality of the fabric by considering the consistency of yarn tension and weaving speed. The calculation of the comprehensive quality monitoring index further integrates the spinning uniformity, tensile shaping stability and fabric structural integrity index, providing a comprehensive quality assessment for the production process, helping enterprises to improve product quality and market competitiveness while maintaining cost-effectiveness. BRIEF DESCRIPTION OF THE DRAWINGS
[0083] Figure 1 A flow chart of a fiber fabric production quality monitoring method provided by an embodiment of the present invention;
[0084] Figure 2 A framework diagram of a fiber fabric production quality monitoring system provided by an embodiment of the present invention;
[0085] Figure 3 A monitoring flow chart of a fiber fabric production quality monitoring system provided by an embodiment of the present invention. DETAILED DESCRIPTION
[0086] The following will be combined with the drawings in the embodiments of the present invention to clearly and completely describe the technical solutions in the embodiments of the present invention. Obviously, the described embodiments are only part of the embodiments of the present invention, not all of the embodiments. Based on the embodiments of the present invention, all other embodiments obtained by ordinary technicians in this field without creative work are within the scope of protection of the present invention.
[0087] The production process of polyester fiber fabrics includes: first, polyethylene terephthalate PET is melted and extruded through a spinning head to form fibers. Then, these fibers are stretched and heat-set to enhance their physical properties. Finally, the fibers are made into fabrics through weaving technology.
[0088] Polyester fiber, mainly referring to polyethylene terephthalate (PET) fiber, is a synthetic fiber made from raw materials such as oil, coal, air and water. This fiber is extremely popular in the textile industry due to its excellent durability, wrinkle resistance and shape retention. In recent years, polyester fiber is mainly made from plastic bottles. This method can convert a large amount of plastic waste into useful products, reducing landfill and environmental pollution.
[0089] It is very important to monitor the production process of polyester fiber fabrics, because this can ensure that the quality of the fabrics meets the standard specifications, reduce production defects and improve production efficiency. The monitoring process can help to timely detect problems on the production line, such as insufficient fiber strength, unstable melt temperature or mechanical failure, so as to take measures to prevent defective products from entering the market. In addition, a good production monitoring system can optimize the use of resources, reduce waste and ensure the sustainability of production.
[0090] Traditional polyester fabric production monitoring methods mainly rely on simple mechanical testing, which includes using a tensile machine to measure the strength and elongation of the fiber, and using tools such as thickness gauges and density gauges to check the physical properties of the fabric. Although this method that relies on direct measurement and observation has been used for decades, it is usually inefficient and difficult to achieve high-precision and real-time production quality control.
[0091] The purpose of the present invention is to provide a fiber fabric production quality monitoring method, which is applied to a fiber fabric production quality monitoring system. The specific method flow chart and system structure diagram refer to Figure 1 and Figure 2 .
[0092] Embodiment 1
[0093] This embodiment analyzes the production process of sample 1;
[0094] Reference Figure 1 S10 in the embodiment, S10 is applied to a melt spinning analysis unit of a fiber fabric production quality monitoring system, including a first production monitoring and a first state monitoring, and is used to perform a first quality monitoring on the melt spinning stage to obtain a spinning uniformity index;
[0095] Further, the first production monitoring includes: obtaining a first image of the fiber after melt spinning, and obtaining a first production quality value by preprocessing and surface analyzing the first image; the first state monitoring is used to monitor melt fluidity and spinning speed during the melt spinning process to obtain a first state quality value; the spinning uniformity index is obtained by comprehensively analyzing the first state quality value and the first production quality value;
[0096] Further, the preprocessing includes graying, denoising and edge detection of the first image;
[0097] The specific analysis process of the surface analysis includes:
[0098] Using the Hough transform algorithm to detect the fiber edge, obtain the fiber diameter; comparing the fiber diameter with the standard diameter, and calculating the difference;
[0099] Using a convolutional neural network to detect appearance defects on the fiber surface; the appearance defects include breakage, agglomeration and foreign matter;
[0100] The defect rate is calculated; the calculation formula of the defect rate is: Among them, S1 is the total area of the defect area, S 总 is the total fiber area;
[0101] The calculation formula of the first production quality value is:
[0102]
[0103] Among them, Q1 is the first production quality value, N1 is the number of monitoring samples of the first production monitoring, and D i is the diameter of the ith fiber, D std It is a standard diameter size;
[0104] The melt fluidity is obtained by a rheometer; the spinning speed is obtained by a speed sensor;
[0105] The calculation formula of the first state quality value is:
[0106]
[0107] Among them, P1 is the first state quality value, S avg ,S std They represent the average and standard melt fluidity, σ v is the standard deviation of spinning speed, V avg is the average spinning speed;
[0108] The formula for the spinning uniformity index is:
[0109] F1=β1·Q1+β2·P1;
[0110] Among them, F1 is the spinning uniformity index; β1, β2 are weight coefficients, and β1+β2=1.
[0111] In this embodiment, some data of the melt spinning stage are shown in Table 1:
[0112] Table 1 Partial data of melt spinning stage
[0113]
[0114] In this embodiment, the fiber fabric production quality monitoring method significantly improves the quality control accuracy and efficiency of the spinning process by performing detailed first production monitoring and first state monitoring of the fibers in the melt spinning stage. The method uses advanced image processing techniques, including graying, denoising and edge detection, as well as Hough transform algorithms and convolutional neural networks to identify and analyze fiber diameters and surface defects, including breakage, agglomeration and foreign matter. Through these techniques, the defect rate and first production quality value of the fiber can be accurately calculated, and the spinning uniformity index can be further generated by combining the monitoring data of the melt fluidity and spinning speed. The application of this comprehensive analysis not only ensures the quality consistency of the fiber, but also provides strong technical support for the timely discovery and resolution of potential problems in the spinning process, thereby ensuring the efficient operation of the entire production process and the high quality standards of the final product.
[0115] Reference Figure 1 S20 in the embodiment, S20 is applied to a stretching and shaping analysis unit of a fiber fabric production quality monitoring system, including a second production monitoring and a second state monitoring, and is used to perform a second quality monitoring on the stretching and shaping stage to obtain a stretching and shaping stability index;
[0116] Further, the second production monitoring includes an online tensile testing device for monitoring the strength and elongation of the fiber during the stretching process to obtain a second production quality value; the second state monitoring includes a thermal imager for monitoring the temperature and time during the heat setting process to obtain a second state quality value; the tensile setting stability index is obtained by comprehensively analyzing the second state quality value and the second production quality value;
[0117] The calculation formula of the second production quality value is:
[0118]
[0119] Where, Q2 is the second production quality value, N2 is the number of monitoring samples for the second production monitoring, and C j ,E j are the fiber strength and fiber elongation of the jth sample, respectively, D std ,E std are standard fiber strength and standard fiber elongation, respectively;
[0120] The calculation formula of the second state quality value is:
[0121]
[0122] Among them, P2 is the second state quality value, T avg ,t std represent the average measured temperature and actual heat setting time, T std ,tstd They are standard temperature and standard heat setting time respectively;
[0123] The formula for the tensile setting stability index is:
[0124] F2=β3·Q2+β4·P2;
[0125] Among them, F2 is the tensile setting stability index; β3, β4 are weight coefficients, and β3+β4=1.
[0126] In this embodiment, some data of the stretching and shaping stage are shown in Table 2:
[0127] Table 2 Partial data table of the stretching and shaping stage
[0128]
[0129] The stretch setting analysis unit plays a key role in the fiber fabric production quality monitoring system. Through the high-precision online tensile testing equipment and thermal imager used in the second production monitoring and second state monitoring, the fiber strength, elongation, and temperature and time of the heat setting process in the stretch setting stage are effectively monitored. This monitoring not only ensures the accuracy and repeatability of the stretch setting process, but also can accurately evaluate the stability of the stretch setting process through the calculated stretch setting stability index, thereby ensuring that the quality of the fiber fabric in this critical production stage meets the preset high standards.
[0130] Reference Figure 1 S30 in which S30 is applied to a textile analysis unit of a fiber fabric production quality monitoring system, including a third production monitoring and a third state monitoring, and is used to perform a third quality monitoring on the textile stage to obtain a fabric structure integrity index;
[0131] Further, the third production monitoring includes acquiring a second image of the fabric, and acquiring a third production quality value by performing defect analysis on the second image; the defect analysis includes analyzing whether the fabric has broken yarn and wrong weaving; the third state monitoring includes a sensor group for monitoring the yarn tension and weaving speed during the textile process to acquire a third state quality value; the sensor group includes a tension sensor and a speed sensor; the fabric structural integrity index is acquired by comprehensively analyzing the third state quality value and the third production quality value;
[0132] The specific process of the defect analysis includes:
[0133] Performing a preprocessing operation on the second image to obtain a third image; the preprocessing operation includes graying, denoising and contrast enhancement;
[0134] Performing feature detection on the third image; the feature detection includes edge detection and texture analysis; the edge detection includes using a Canny edge detection algorithm to identify yarn edges and obtain edge features; the texture analysis includes analyzing a texture pattern of the fabric through a gray-level co-occurrence matrix and obtaining texture features;
[0135] The edge features and the texture features are analyzed using a CNN model to obtain specific locations and types of defects; the types include broken yarn and mis-weaving;
[0136] The calculation formula of the third production quality value is:
[0137]
[0138] Among them, δ break ,δ miss are the weights of yarn breakage and misweaving, respectively, severity,break ,λ severity,miss are the severity factors of yarn breakage and misweaving, A total is the total area analyzed, A break A is the total area of the detected broken yarn defect area, miss is the total area of the misweaving defect region detected;
[0139] In this embodiment, the sample 1 is a piece of fabric with an area of 10,000 square millimeters. Through image analysis, a broken yarn defect of 150 square millimeters and a misweaving defect of 50 square millimeters are identified. The weight of the broken yarn is set to 0.6, the weight of the misweaving is set to 0.4, and the severity factor of the broken yarn is set to 1.5, and the severity factor of the misweaving is set to 1.2;
[0140] The third production quality value is 0.9841;
[0141] The calculation formula of the third state quality value is:
[0142]
[0143] Among them, P3 is the third state quality value, L avg ,VS avg Represent the average values of tension and weaving speed, l n ,l vs represent the standard deviation of tension data and weaving speed data respectively;
[0144] The formula for the fabric structural integrity index is:
[0145] F3=β5·Q3+β6·P3;
[0146] Wherein, F3 is the fabric structural integrity index; β5, β6 are weight coefficients, and β5+β6=1.
[0147] The textile analysis unit uses advanced image processing and sensor technology to conduct comprehensive quality monitoring of fabrics in the textile stage, effectively improving the ability to detect the structural integrity of fabrics. By pre-processing the second image of the fabric by graying, denoising and contrast enhancement, combined with the Canny edge detection algorithm and grayscale co-occurrence matrix analysis, this system can accurately identify and locate broken yarns and mis-weaving defects in the fabric. In addition, the use of tension sensors and speed sensors to monitor the yarn tension and weaving speed during the textile process further enhances the control of fabric quality.
[0148] Reference Figure 1 S40 in the invention is applied to a comprehensive analysis unit of a fiber fabric production quality monitoring system to obtain a comprehensive quality monitoring index; the comprehensive quality monitoring index is obtained by comprehensively analyzing a spinning uniformity index, a stretching and shaping stability index, and a fabric structure integrity index.
[0149] Furthermore, the calculation formula of the comprehensive quality monitoring index is:
[0150] F al =α1·F1+α2·F2+α3·F3;
[0151] Among them, F al is a comprehensive quality monitoring index, F1, F2, F3 are spinning uniformity index, stretching setting stability index and fabric structure integrity index respectively, α1, α2, α3 are the weights of spinning uniformity index, stretching setting stability index and fabric structure integrity index respectively, and α1>α2>α3.
[0152] In the calculation of the comprehensive quality monitoring index, different weights are assigned to each index to reflect their degree of influence and importance on the quality of the final product. The spinning uniformity index is given the highest weight because the spinning process is the basic stage of fiber manufacturing, and its uniformity directly affects the effect of all subsequent processing steps and the quality of the finished product. If the fiber uniformity during the spinning process is poor, it will lead to problems in the stretching, shaping and spinning processes that cannot be effectively remedied, thus affecting the stability of the entire production chain and the consistency of the product. On the contrary, the fabric structural integrity index, although important, is given the lowest weight because even if there are slight unevenness or defects in the structural integrity of the fabric, it is still possible to adjust or correct it through subsequent processing or finishing steps. Therefore, although the integrity of the fabric structure is a key factor affecting the appearance and performance of the final product, its weight is relatively low in order to emphasize the core role of spinning uniformity in ensuring production quality.
[0153] The comprehensive analysis unit is applied to the fiber fabric production quality monitoring system to effectively summarize and evaluate the key quality indicators of each production stage, including the spinning uniformity index, the stretching stability index and the fabric structure integrity index, thereby generating a comprehensive quality monitoring index. The index comprehensively reflects the quality level of the entire production process by weighted analysis of the above indicators, allowing production managers to obtain a comprehensive quality assessment result. The use of this system has greatly improved the ability to control production quality and ensured the consistency and reliability of products.
[0154] This embodiment obtains a monitoring flow chart of a fiber fabric production quality monitoring system according to the above steps. Figure 3 . The specific monitoring steps include: first, inputting PET into the melt spinning processing module to obtain fiber; performing the first production monitoring and the first state monitoring on the melt spinning stage by image processing technology, convolutional neural network and sensor technology, obtaining the first production quality value and the first state quality value, and obtaining the spinning uniformity index by comprehensive analysis of the first production quality value and the first state quality. Then, inputting the fiber into the stretching and shaping processing module. Obtaining the stretched fiber; performing the second production monitoring and the second state monitoring on the stretching and shaping stage by online tensile testing equipment and thermal imager, obtaining the second production quality value and the second state quality value, and obtaining the stretching and shaping stability index by comprehensive analysis of the second production quality value and the second state quality; further, inputting the stretched fiber into the textile processing module to obtain the fiber fabric; performing the third production monitoring and the third state monitoring on the textile stage by image processing technology and sensor technology, obtaining the third production quality value and the third state quality value, and obtaining the fabric structure integrity index by comprehensive analysis of the third production quality value and the third state quality; finally, obtaining the comprehensive quality monitoring index by comprehensive analysis of the spinning uniformity index, the stretching and shaping stability index and the fabric structure integrity index.
[0155] A fiber fabric production quality monitoring method proposed in this embodiment can comprehensively improve the monitoring efficiency and quality accuracy of the entire fiber fabric production process. Through systematic monitoring and data analysis of key production stages such as melt spinning, stretching and shaping, and weaving, this method can track the production quality of each step in detail. From the spinning uniformity index in the melt spinning stage, to the stability index in the stretching and shaping stage, to the structural integrity index in the weaving stage, each index is obtained through precise real-time data collection and analysis, ensuring quality control of each link in the production process. Finally, these indexes are combined to form a comprehensive quality monitoring index, providing production managers with a quantitative and clear quality assessment index. This comprehensive quality monitoring method not only optimizes the production process and reduces the output of defective products, but also significantly improves the market competitiveness of the final product, meeting the strict requirements of the high-quality fiber fabric market.
[0156] Embodiment 2
[0157] This embodiment analyzes the generation process of sample 2 based on the description of a fiber fabric production quality monitoring method and system in embodiment 1.
[0158] In this embodiment, the sample 2 is a piece of fabric with an area of 12,000 square millimeters. Through image analysis, a broken yarn defect of 200 square millimeters and a misweaving defect of 100 square millimeters are identified. The weight of the broken yarn is set to 0.7, the weight of the misweaving is set to 0.3, and the severity factor of the broken yarn is set to 1.8, and the severity factor of the misweaving is set to 1.1;
[0159] The third production quality value is 0.97618;
[0160] In this embodiment, the analytic hierarchy process is used to determine the weights for the spinning uniformity index, the stretching stability index and the fabric structure integrity index in the comprehensive quality monitoring index. The specific steps include:
[0161] Define goals and criteria, including spinning uniformity index, stretch setting stability index and fabric structural integrity index;
[0162] Constructing the judgment matrix,By analyzing the historical data, a judgment matrix table as shown in Table 3 is constructed, where each element represents the importance of the row factor relative to the column factor.
[0163] Table 3 Judgment matrix
[0164] Spinning uniformity Tensile setting stability Fabric structural integrity Spinning uniformity 1 3 5 Tensile setting stability 1 / 3 1 3 Fabric structural integrity 1 / 5 1 / 3 1
[0165] Standardize the judgment matrix table and calculate the average value of each row as the weight;
[0166] In this embodiment, the calculated weights are: spinning uniformity: 0.60; stretching setting stability: 0.30; fabric structural integrity: 0.10;
[0167] In order to verify that the judgment matrix has good consistency, in this embodiment, the consistency index (CI) and the random consistency index (RI) are calculated, and the consistency ratio (CR) is obtained through CI and RI. It is determined whether the CR is less than 0.1 to ensure the consistency acceptance of the judgment matrix.
[0168] The calculation formula of CI is:
[0169]
[0170] Where n is the dimension of the judgment matrix; λ max is the maximum eigenvalue; for a completely consistent judgment matrix, λ max =n;
[0171] λ max The acquisition of λ includes multiplying all elements of each row of the judgment matrix and taking the nth root, and finally adding these values to obtain λ max In this embodiment, λ max =3.05;
[0172] RI is derived from the average consistency index of the randomly generated matrix. In this embodiment, this value is pre-calculated and is 0.58;
[0173] Furthermore, the calculation formula of CR is:
[0174]
[0175] In this embodiment, CI is 0.025 and CR is 0.043; because the CR value is less than 0.1, the judgment matrix constructed in this embodiment has good consistency and the analysis results can be accepted;
[0176] Although embodiments of the present invention have been shown and described, it will be appreciated by those skilled in the art that various changes, modifications, substitutions and variations may be made to the embodiments without departing from the principles and spirit of the present invention, and that the scope of the present invention is defined by the appended claims and their equivalents.
Claims
1. A fiber fabric production quality monitoring method, characterized in that: include: Performing a first quality monitoring of the melt spinning stage to obtain a spinning uniformity index; The first quality monitoring includes first production monitoring and first status monitoring; The first production monitoring includes: obtaining a first image of the fiber after melt spinning, and obtaining a first production quality value by preprocessing and surface analysis of the first image; the first state monitoring is used to monitor melt fluidity and spinning speed during the melt spinning process to obtain a first state quality value; the spinning uniformity index is obtained by comprehensively analyzing the first state quality value and the first production quality value; Performing a second quality monitoring on the stretching and shaping stage to obtain a stretching and shaping stability index; the second quality monitoring includes a second production monitoring and a second state monitoring; Performing a third quality monitoring on the weaving stage to obtain a fabric structural integrity index; the third quality monitoring includes a third production monitoring and a third status monitoring; A comprehensive quality monitoring index is obtained by comprehensively analyzing the spinning uniformity index, the stretching setting stability index and the fabric structure integrity index.
2. A fiber fabric production quality monitoring method according to claim 1, characterized in that: The preprocessing includes gray-scaling, denoising and edge detection on the first image; The specific analysis process of the surface analysis includes: Using the Hough transform algorithm to detect the fiber edge, the fiber diameter is obtained; the fiber diameter is compared with the standard diameter, and the difference is calculated; Using a convolutional neural network to detect appearance defects on the fiber surface; the appearance defects include breakage, agglomeration, and foreign matter; The defect rate is calculated; the calculation formula of the defect rate is: Among them, S1 is the total area of the defect area, S 总 is the total fiber area; The calculation formula of the first production quality value is: Among them, Q1 is the first production quality value, N1 is the number of monitoring samples of the first production monitoring, D i is the diameter of the i-th fiber, D std For standard diameter size; The melt fluidity is obtained by a rheometer; the spinning speed is obtained by a speed sensor; The calculation formula of the first state quality value is: Among them, P1 is the first state quality value, S avg ,S std They represent the average value and standard melt fluidity of the melt, σ v is the standard deviation of spinning speed, V avg is the average spinning speed; The formula for the spinning uniformity index is: F1=β1·Q1+β2·P1; Wherein, F1 is the spinning uniformity index; β1, β2 are weight coefficients, and β1+β2=1.
3. The fiber fabric production quality monitoring method according to claim 1, characterized in that: The second production monitoring includes an online tensile testing device for monitoring the strength and elongation of the fiber during the stretching process to obtain a second production quality value; The second state monitoring includes a thermal imager for monitoring the temperature and time during the heat setting process to obtain a second state quality value; Obtaining the stretch setting stability index by comprehensively analyzing the second state quality value and the second production quality value; The calculation formula of the second production quality value is: Where, Q2 is the second production quality value, N2 is the number of monitoring samples for the second production monitoring, and C j ,E j are the fiber strength and fiber elongation of the jth sample, D std ,E std are standard fiber strength and standard fiber elongation, respectively; The calculation formula of the second state quality value is: Among them, P2 is the second state quality value, T avg ,t std Represent the average measured temperature and actual heat setting time, T std ,t std are standard temperature and standard heat setting time respectively; The formula for the tensile setting stability index is: F2=β3·Q2+β4·P2; Among them, F2 is the tensile setting stability index; β3 and β4 are weight coefficients, and β3+β4=1.
4. The fiber fabric production quality monitoring method according to claim 1, characterized in that: The third production monitoring includes acquiring a second image of the fabric and performing defect analysis on the second image to acquire a third production quality value; the defect analysis includes analyzing whether the fabric has broken yarns and misweaving; The third state monitoring includes a sensor group for monitoring the yarn tension and weaving speed during the weaving process to obtain the third state quality value; the sensor group includes a tension sensor and a speed sensor; Obtaining the fabric structural integrity index by comprehensively analyzing the third state quality value and the third production quality value; The specific process of the defect analysis includes: Performing a preprocessing operation on the second image to obtain a third image; the preprocessing operation includes grayscale conversion, denoising, and contrast enhancement; Performing feature detection on the third image; the feature detection includes edge detection and texture analysis; the edge detection includes using a Canny edge detection algorithm to identify yarn edges and obtain edge features; the texture analysis includes analyzing the texture pattern of the fabric through a gray level co-occurrence matrix and obtaining texture features; Analyzing the edge features and the texture features using a CNN model to obtain specific locations and types of defects, including yarn breakage and misweaving; The calculation formula of the third production quality value is: Among them, δ break ,δ miss are the weights of yarn breakage and misweaving, λ severity,break ,λ severity,miss are the severity factors of yarn breakage and misweaving, A total is the total area analyzed, A break A is the total area of the detected broken yarn defect area, miss is the total area of the misweaving defect regions detected; The calculation formula of the third state quality value is: Among them, P3 third state quality value, L avg ,VS avg Represent the average values of tension and weaving speed, l n ,l vs represent the standard deviation of tension data and weaving speed data respectively; The formula for the fabric structural integrity index is: F3=β5·Q3+β6·P3; Wherein, F3 is the fabric structural integrity index; β5, β6 are weight coefficients, and β5+β6=1.
5. The fiber fabric production quality monitoring method according to claim 1, characterized in that: The calculation formula of the comprehensive quality monitoring index is: F al =α1·F1+α2·F2+α3·F3; Among them, F al is a comprehensive quality monitoring index, F1, F2, F3 are the spinning uniformity index, stretching stability index and fabric structure integrity index respectively, α1, α2, α3 are the weights of the spinning uniformity index, stretching stability index and fabric structure integrity index respectively, and α1>α2>α3.
6. A fiber fabric production quality monitoring system, characterized in that: include: a melt spinning analysis unit, comprising a first production monitoring unit and a first state monitoring unit, for performing a first quality monitoring on the melt spinning stage and obtaining a spinning uniformity index; The stretching and shaping analysis unit includes a second production monitoring unit and a second state monitoring unit, which is used to perform a second quality monitoring on the stretching and shaping stage and obtain a stretching and shaping stability index; The textile analysis unit includes a third production monitoring unit and a third state monitoring unit, which is used to perform third quality monitoring of the textile stage and obtain the fabric structural integrity index; The comprehensive analysis unit is used to obtain a comprehensive quality monitoring index; the comprehensive quality monitoring index is obtained by comprehensively analyzing the spinning uniformity index, the stretching and setting stability index and the fabric structure integrity index.
7. A fiber fabric production quality monitoring system according to claim 6, characterized in that: The first production monitoring includes: obtaining a first image of the fiber after melt spinning, and obtaining a first production quality value by preprocessing and surface analyzing the first image; the first state monitoring is used to monitor the melt fluidity and spinning speed during the melt spinning process to obtain a first state quality value.
8. A fiber fabric production quality monitoring system according to claim 7, characterized in that: The preprocessing includes gray-scaling, denoising and edge detection on the first image; The specific analysis process of the surface analysis includes: Using the Hough transform algorithm to detect the fiber edge, the fiber diameter is obtained; the fiber diameter is compared with the standard diameter, and the difference is calculated; Using a convolutional neural network to detect appearance defects on the fiber surface; the appearance defects include breakage, agglomeration, and foreign matter; The defect rate is calculated; the calculation formula of the defect rate is: Among them, S1 is the total area of the defect area, S 总 is the total fiber area; The calculation formula of the first production quality value is: Among them, Q1 is the first production quality value, N1 is the number of monitoring samples of the first production monitoring, D i is the diameter of the i-th fiber, D std For standard diameter size; The melt fluidity is obtained by a rheometer; the spinning speed is obtained by a speed sensor; The calculation formula of the first state quality value is: Among them, P1 is the first state quality value, S avg ,S std They represent the average value and standard melt fluidity of the melt, σ v is the standard deviation of spinning speed, V avg is the average spinning speed.
9. A fiber fabric production quality monitoring system according to claim 6, characterized in that: The formula for the spinning uniformity index is: F1=β1·Q1+β2·P1; Wherein, F1 is the spinning uniformity index; β1, β2 are weight coefficients, and β1+β2=1.
10. The fiber fabric production quality monitoring system according to claim 6, characterized in that: The second production monitoring includes an online tensile testing device for monitoring the strength and elongation of the fiber during the stretching process to obtain a second production quality value; The second state monitoring includes a thermal imager for monitoring the temperature and time during the heat setting process to obtain a second state quality value; The stretch setting stability index is obtained by comprehensively analyzing the second state quality value and the second production quality value.
Citation Information
Cited By
Cotton cloth design automatic evaluation system
CN119904143A
Production management method and system of high-elastic polyester fabric and storage medium
CN119963496A
Production management method, system and storage medium for high-elastic polyester fabric
CN119963496B
