Real-time quality rating method and system for yarn spinning

Abstract

The invention relates to the field of yarn spinning, in particular to a real-time quality rating method and system for yarn spinning. The method comprises the following steps: acquiring a space-time charge data set, and generating a yarn electrostatic state data set through a multi-mode sensing fusion strategy; on the basis of the yarn static state data set, yarn spinning quality defects are recognized by combining yarn form data collected in real time, two-dimensional dynamic rating is executed according to defect types and severity, and yarn real-time quality rating is generated; according to the yarn real-time quality rating, carrying out change trend analysis on the yarn real-time quality rating to generate a yarn subsequent quality change trend; and executing a corresponding process regulation and control instruction according to the subsequent quality change trend of the yarn, and outputting a quality rating log. In the yarn spinning process, process look-ahead optimization is achieved, quality tracing is perfected, and the accuracy and real-time performance of quality rating are improved.

Description

Technical Field

[0001] The present application relates to the field of yarn textiles, and in particular to a real-time quality rating method and system for yarn textiles. Background Art

[0002] In the textile industry, yarn, as the basic unit of fabric production, and its quality stability constitute the core lifeline of the trillion-level clothing, home textile and industrial textile manufacturing system, and are the foundation for supporting the efficient operation and value realization of the huge textile industry chain.

[0003] However, existing yarn quality detection technology is unable to accurately detect and quantify the dynamic details of high-speed yarns during the production process, resulting in difficulty in timely intervention in quality fluctuations, continued output of defective products, and increased waste of raw materials, ultimately weakening product competitiveness and raising production costs. Summary of the Invention

[0004] The present application provides a real-time quality rating method and system for yarn textiles to solve the above-mentioned technical problems.

[0005] In a first aspect, the present application provides a real-time quality rating method for yarn textiles, the method comprising: A spatiotemporal charge dataset is acquired, and a yarn electrostatic state dataset is generated through a multimodal sensing fusion strategy; based on the yarn electrostatic state dataset and combined with real-time collected yarn morphology data, yarn textile quality defects are identified, and a two-dimensional dynamic rating is performed according to the defect type and severity to generate a real-time yarn quality rating; based on the real-time yarn quality rating, a change trend analysis is performed on the real-time yarn quality rating to generate a subsequent quality change trend of the yarn; based on the subsequent quality change trend of the yarn, corresponding process control instructions are executed, and a quality rating log is output.

[0006] This solution first collects spatiotemporal charge data, then generates a yarn electrostatic state dataset through multimodal sensor fusion, achieving precise quantification of electrostatic factors. The electrostatic state data is then fused with morphological data to extract defect-related electrostatic representations. Real-time quality results are generated through two-dimensional dynamic rating, improving the comprehensiveness and accuracy of defect identification. Combining yarn category information and electrostatic association weights, the electrostatic data is calibrated and quality change trends are predicted, enhancing the pertinence of the analysis. Proactive process control is performed based on trends, and a quality rating log is output to the user, forming a closed-loop control system. By generating an electrostatic state dataset through multimodal sensor fusion, identifying defects and performing two-dimensional ratings based on morphological data, and then predicting and regulating trends based on category information and electrostatic weights, the dynamic details of high-speed yarns can be accurately detected and quantitatively rated in real time, enabling timely intervention in quality fluctuations, reducing the continuous output of defective products, and avoiding increased waste of raw materials, ultimately enhancing product competitiveness and reducing production costs.

[0007] Optionally, the multimodal sensing fusion strategy includes: utilizing a spatially deployed annular array of electrostatic sensors to collect a spatiotemporal charge dataset on key points of the yarn in real-time and non-contact manner; the spatiotemporal charge dataset includes a real-time charge intensity value and a real-time charge distribution characteristic; utilizing a high-frame rate visual sensor to capture yarn morphology data and motion trajectory in real time; synchronously analyzing the spatiotemporal correlation between the real-time charge intensity value, the real-time charge distribution characteristic and the yarn morphology data to identify abnormal electrostatic adsorption trajectories caused by quality defects; and generating the yarn electrostatic state dataset based on the spatiotemporal charge dataset, the yarn morphology data and the abnormal electrostatic adsorption trajectories.

[0008] Optionally, based on the yarn electrostatic state data set and combined with the yarn morphology data collected in real time, the yarn textile quality defects are identified, and a two-dimensional dynamic rating is performed according to the defect type and severity to generate an initial yarn quality rating, including: based on the yarn electrostatic state data set and the yarn morphology data collected in real time, the yarn defect type is identified, and the severity of the dominant defect corresponding to the yarn defect type is determined; the yarn defect types include excessive hairiness, uneven yarn and fiber knotting; several of the yarn defect types are normalized to a unified numerical range to determine the normalized yarn defect type; a dynamic rating score is determined according to the normalized yarn defect type and the severity of the dominant defect; based on the dynamic rating score, a real-time yarn quality rating is generated, and the real-time yarn quality rating is divided into three levels: excellent, medium and poor.

[0009] Optionally, the method of identifying the type of yarn defect based on the yarn electrostatic state data set and the yarn morphology data collected in real time, and determining the severity of the dominant defect corresponding to the yarn defect type, includes: acquiring the yarn surface image and the corresponding real-time charge intensity value in real time, and performing the following hairiness density grid analysis step according to the yarn surface image: when the number of discrete fibers in several consecutive detection data exceeds a preset first density threshold and the average value of their electrostatic intensity is greater than a preset multiple of the number of visual hairiness, it is determined to be a local hairiness aggregation defect; when the number of discrete fibers in a unit detection area exceeding a preset proportion exceeds a preset second density threshold and the overall electrostatic distribution unevenness is greater than a preset diffuse hairiness ratio, it is determined to be a diffuse hairiness excess defect; according to the combination pattern of the number of local aggregation areas and the diffuse distribution ratio, combined with the coupling coefficient of the electrostatic intensity increase and the hairiness density, the spatial distribution state of the hairiness is analyzed to quantify the degree of dominance of the hairiness density.

[0010] Optionally, the yarn defect type is identified based on the yarn electrostatic state data set and the yarn morphology data collected in real time, and the severity of the dominant defect corresponding to the yarn defect type is determined, including: real-time acquisition of the yarn axial image sequence, and execution of the following three-dimensional yarn fluctuation analysis steps based on the yarn axial image sequence: marking the diameter abnormal area where the diameter difference between adjacent sampling points exceeds the diameter tolerance threshold as a sudden uneven area, and synchronously detecting the angular change rate of the electrostatic adsorption trajectory of the abnormal diameter area, and marking the current diameter abnormal area as an electrostatic turbulence when the angular change rate exceeds a preset turbulence threshold. The flow pattern suddenly changes unevenly; the trend abnormal area corresponding to the unidirectional continuous increase / decrease trend of the diameter that exceeds the preset number of points continuously is marked as a gradual uneven area, and the electrostatic gradient distribution slope of the gradual area is obtained simultaneously. When the slope is in the same direction as the diameter change trend and exceeds the associated threshold, it is marked as electrostatic cumulative gradual unevenness; the proportion of the length of the sudden uneven area and the length of the gradual uneven area in the unit detection area is determined, and the degree of the unevenness dominance of the strip is quantified by combining the maximum diameter deviation value and the electrostatic disorder intensity parameter that are positively correlated with the degree of unevenness dominance of the strip, and the electrostatic gradient consistency parameter that is negatively correlated with the degree of unevenness dominance of the strip.

[0011] Optionally, the method of identifying the type of yarn defect based on the yarn electrostatic state data set and the yarn morphology data collected in real time, and determining the severity of the dominant defect corresponding to the yarn defect type, includes: acquiring yarn surface scanning imaging data in real time, and synchronously capturing the corresponding electrostatic intensity extreme value, and performing the following impurity multi-scale feature extraction steps to clarify the impurity attributes: when there is a dark clumping structure with irregular boundaries whose charge decay rate is less than a preset decay rate threshold, the dark clumping structure is determined to be a fiber agglomerate; when there are reflective particles with geometric edges whose electrostatic intensity extreme value is greater than a preset multiple of the particle size, the reflective particles are determined to be hard impurities; when there is a radial fiber entanglement structure whose electrostatic intensity is in an oscillating and fluctuating state, the radial fiber entanglement structure is determined to be a yarn defect nodule; according to the size grade, optical contrast and boundary clarity of the impurity multi-scale features, the size grade, the optical contrast and the boundary clarity are weighted according to the impurity attributes to quantify the degree of dominant fiber knots.

[0012] Optionally, the real-time quality rating of the yarn is generated according to the dynamic rating score, and the real-time quality rating of the yarn is divided into three grades: excellent, medium and poor, including: based on a preset rule table, mapping the identified several yarn defect types to defect type identifiers, and mapping the assessed severity of the dominant defect to a severity level identifier; performing string splicing on several defect type identifiers and corresponding severity level identifiers, and using the spliced ​​result as a unit rating code; after all the unit rating codes are spliced, performing secondary string splicing on several unit rating codes to generate a composite rating code; according to the preset mapping rules, mapping the composite rating code to the initial quality grade of excellent / medium / poor; the mapping The rules include: determining the total number of all defects as the defect density, and identifying the highest severity level identifier in the composite rating code; if the defect density is greater than a first density threshold or the highest severity level identifier reaches the first severity level threshold, it is judged as poor; if the defect density does not meet the aforementioned conditions for being judged as poor but is greater than a second density threshold or the highest severity level identifier reaches the second severity level threshold, it is judged as medium; if the defect density does not exceed the second density threshold and the highest severity level identifier does not reach the second severity level threshold, it is judged as excellent; wherein, the first density threshold is greater than the second density threshold, and the first severity level threshold is higher than the second severity level threshold.

[0013] Optionally, the yarn real-time quality rating is analyzed for a change trend based on the yarn real-time quality rating to generate a subsequent quality change trend of the yarn, including: extracting trajectory time series dynamic features including trajectory fluctuation frequency, directional consistency and intensity variation amplitude based on the electrostatic adsorption abnormal trajectory; identifying trajectory evolution patterns based on the dynamic features: when the trajectory fluctuation frequency is lower than a preset low-frequency threshold and the directional consistency is higher than a stable threshold, it is a steady slowdown mode; when the trajectory fluctuation frequency is higher than a preset high-frequency threshold and the intensity variation amplitude continues to increase, it is a deterioration diffusion mode; when the directional consistency is lower than a disorder threshold and the intensity variation amplitude periodically decreases, it is an intermittent oscillation mode; mapping the trajectory evolution pattern to a quality trend: the steady slowdown mode is mapped to a defect alleviation trend, the deterioration diffusion mode is mapped to a defect escalation trend, and the intermittent oscillation mode is mapped to a quality fluctuation trend; and generating a subsequent quality change trend of the yarn based on the quality trend combined with the severity of the dominant defect.

[0014] Optionally, according to the subsequent quality change trend of the yarn, corresponding process control instructions are executed, and a quality rating log is output, including: performing differentiated control according to the type of subsequent quality change trend of the yarn: if it is determined to be the defect escalation trend, then triggering linkage control, reducing the spinning speed and increasing the environmental humidification; if it is determined to be the defect alleviation trend, then maintaining the current process and reducing the detection density; if it is determined to be the quality fluctuation trend, then dynamically adjusting the environmental temperature and humidity and directionally starting static electricity neutralization; associating the real-time quality rating of the yarn, the subsequent quality change trend of the yarn and the control instructions with the production line batch and time, and generating and outputting a quality rating log.

[0015] In a second aspect, the present application provides a real-time quality rating system for yarn weaving, the system comprising: The charge monitoring module is used to obtain the spatiotemporal charge data set and generate the yarn electrostatic state data set through a multimodal sensor fusion strategy; A two-dimensional rating module is used to identify yarn textile quality defects based on the yarn electrostatic state data set and the real-time collected yarn morphology data, and perform two-dimensional dynamic rating according to the defect type and severity to generate a real-time yarn quality rating; A trend deduction module is used to analyze the change trend of the yarn real-time quality rating according to the yarn real-time quality rating, and generate the subsequent quality change trend of the yarn; The control execution module is used to execute corresponding process control instructions according to the subsequent quality change trend of the yarn and output a quality rating log. BRIEF DESCRIPTION OF THE DRAWINGS

[0016] In order to more clearly illustrate the embodiments of the present application or the technical solutions in the prior art, a brief introduction will be given below to the drawings required for use in the embodiments or the description of the prior art. Obviously, the drawings described below are some embodiments of the present application. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative labor.

[0017] Figure 1 A schematic diagram of an application scenario provided in one embodiment of the present application; Figure 2 A flowchart of a real-time quality rating method for yarn textiles provided in one embodiment of the present application; Figure 3 A schematic structural diagram of a real-time quality rating system for yarn textiles provided in one embodiment of the present application. DETAILED DESCRIPTION

[0018] To make the purpose, technical solutions, and advantages of the embodiments of the present application clearer, the technical solutions in the embodiments of the present application will be clearly and completely described below in conjunction with the drawings in the embodiments of the present application. Obviously, the described embodiments are part of the embodiments of the present application, not all of the embodiments. Based on the embodiments in the present application, all other embodiments obtained by ordinary technicians in this field without making creative efforts are within the scope of protection of this application.

[0019] In this document, the term "and / or" simply describes a relationship between related objects, indicating that three possible relationships exist. For example, "A and / or B" can represent: A exists alone, A and B exist simultaneously, or B exists alone. Furthermore, the character " / " in this document, unless otherwise specified, generally indicates an "or" relationship between the related objects.

[0020] The embodiments of the present application are described in further detail below with reference to the accompanying drawings.

[0021] However, existing yarn quality detection technology is unable to accurately detect and quantify the dynamic details of high-speed yarns during the production process, resulting in difficulty in timely intervention in quality fluctuations, continued output of defective products, and increased waste of raw materials, ultimately weakening product competitiveness and raising production costs.

[0022] Based on this, the present application provides a real-time quality rating method and system for yarn textiles. First, spatiotemporal charge data is collected, and a yarn electrostatic state data set is generated through multimodal sensor fusion to achieve accurate quantification of electrostatic factors; the electrostatic state data and morphological data are integrated to extract the electrostatic representation associated with the defects, and real-time quality results are generated through two-dimensional dynamic rating to improve the comprehensiveness and accuracy of defect identification; the yarn category information and electrostatic association weights are combined to calibrate the electrostatic data and predict the quality change trend to enhance the pertinence of the analysis; forward-looking process control is performed based on the trend, and the quality rating log is output to the user to form a closed-loop control. By generating an electrostatic state data set through multimodal sensor fusion, identifying defects and rating them in two dimensions based on morphological data, and then predicting and regulating trends based on category information and electrostatic weights, the dynamic details of high-speed running yarns can be accurately detected and quantitatively rated in real time, so that quality fluctuations can be intervened in a timely manner, the continuous output of defective products can be reduced, and the waste of raw materials can be avoided, ultimately enhancing product competitiveness and reducing production costs.

[0023] Figure 1 This is a schematic diagram of an application scenario provided by this application. This application achieves forward-looking process optimization and improves quality traceability in the yarn spinning process, enhancing the accuracy and real-time performance of quality ratings.

[0024] Specifically, the method of the present application is applied to any server that communicates with an electrostatic induction ring array and obtains a spatiotemporal charge data set provided by the electrostatic induction ring array through the server. First, the spatiotemporal charge data is collected, and a yarn electrostatic state data set is generated through multimodal sensor fusion to achieve accurate quantification of electrostatic factors; the electrostatic state data and morphological data are integrated to extract electrostatic characterizations associated with defects, and real-time quality results are generated through two-dimensional dynamic rating to improve the comprehensiveness and accuracy of defect identification; the yarn category information and electrostatic association weights are combined to calibrate the electrostatic data and predict the quality change trend to enhance the pertinence of the analysis; based on the trend, forward-looking process control is performed, and a quality rating log is output to the user to form a closed-loop control. The specific implementation method can be referred to the following embodiments.

[0025] Figure 2 This is a flowchart of a real-time quality rating method for yarn weaving provided by an embodiment of the present application. The method of this embodiment can be applied to the server in the above scenario. Figure 2 As shown, the method includes: S201. Acquire a spatiotemporal charge dataset and generate a yarn electrostatic state dataset through a multimodal sensing fusion strategy.

[0026] A spatiotemporal charge dataset can refer to a collection of electrostatic charge data collected over time and at different spatial locations during the yarn production process. This data includes real-time charge intensity values ​​and charge distribution characteristics. This data is derived from a ring array of electrostatic sensors surrounding the yarn. A multimodal sensor fusion strategy can be a processing strategy that integrates different types of sensor data from the ring array of electrostatic sensors and high-frame-rate vision sensors.

[0027] The yarn electrostatic state dataset can be a dataset used to characterize the charge data, morphological data and electrostatic adsorption abnormal trajectory of the yarn during the textile process, and can intuitively reflect the electrostatic characteristics of the yarn.

[0028] Specifically, in the yarn weaving process, static electricity is a key hidden factor affecting quality, yet it has long been overlooked by traditional quality inspection methods. Existing technologies mostly rely on manual visual inspection or single-form sensors, which can only capture macroscopic appearance defects of yarn (such as yarn breakage) and lack the ability to accurately monitor and quantitatively evaluate the microscopic dynamics of high-speed yarns in real time. This creates blind spots in the production process, making it difficult to correct quality anomalies in a timely manner, increasing defective product rates, and increasing raw material losses, ultimately hindering product market competitiveness and significantly increasing overall costs. Furthermore, data collected by a single charge sensor is susceptible to interference from environmental temperature and humidity, equipment vibration, and other factors, resulting in one-sided data that cannot accurately reflect the true state of yarn static electricity. This step deploys devices such as a ring array of static sensors at key locations on the textile machine to collect data such as charge distribution and motion trajectory on the yarn surface and surrounding space in real time, forming a spatiotemporal charge dataset. A multimodal sensor fusion strategy is used to extract static characteristic parameters and generate a yarn static state dataset. The acquisition of spatiotemporal charge data sets can comprehensively capture the dynamic changes in charge on the yarn surface and the surrounding space. The multimodal sensing fusion strategy can eliminate the errors of a single sensor and filter out noise interference by integrating multiple sources such as charge and optical data. The final generated yarn electrostatic state data set can accurately quantify the electrostatic properties and provide a reliable basis for subsequent quality analysis.

[0029] S202. Based on the yarn electrostatic state dataset and in combination with the real-time collected yarn morphology data, yarn textile quality defects are identified, and a two-dimensional dynamic rating is performed according to the defect type and severity to generate a real-time yarn quality rating.

[0030] Yarn morphology data can refer to a data set that reflects the physical appearance and structural characteristics of yarns, including parameters such as yarn diameter, hairiness length and density, and variations in yarn thickness. Yarn textile quality defects can refer to problems in the yarn during the textile process that fail to meet quality standards due to factors such as raw material characteristics, equipment parameters, and environmental conditions. Common types include excessive hairiness, uneven yarn thickness, and fiber knots. Two-dimensional dynamic rating can be a method for real-time assessment of yarn quality based on the two dimensions of "defect type" and "severity." The defect type dimension categorizes different quality issues, and the severity dimension categorizes the defects based on their impact on subsequent processing and product performance (e.g., minor, moderate, severe). Yarn real-time quality rating can be based on real-time collected electrostatic state and morphology data. The evaluation results of the current yarn quality level generated by two-dimensional dynamic rating directly reflect the immediate quality status of the yarn.

[0031] Specifically, traditional quality rating methods have two core flaws: First, they rely solely on macroscopic morphology to assess quality, using macroscopic morphological data as the sole basis. This can lead to potential microscopic defects caused by static electricity (such as fiber disarray caused by static electricity) being overlooked. Second, their rating dimensions are limited, judging quality solely through a "pass / fail" rating or simple score. This fails to distinguish defect types (such as the difference between hairiness and yarn breakage) and severity (such as the difference between mild hairiness and the impact of extensive hairiness), resulting in a lack of targeted production control. This step, based on a dataset of yarn electrostatic states and combined with yarn morphological data (diameter, twist, hairiness, etc.) collected in real time by image recognition equipment, uses correlation analysis to identify electrostatic characteristics strongly associated with quality defects (such as the electrostatic distribution pattern when hairiness exceeds the standard). Based on the identified defect type (such as uneven yarn length and yarn breakage) and severity (such as defect proportion and impact range), a dynamic rating is performed along both the "type" and "severity" dimensions to generate a real-time yarn quality rating. By integrating the yarn electrostatic state dataset and morphological data, we can discover the strong correlation characteristics of "static electricity-defect" (such as the periodic fluctuation of static electricity distribution when the yarn is uneven), so that defect identification can be deepened from surface observation to mechanism analysis, greatly improving the recognition accuracy; at the same time, the two-dimensional dynamic rating not only clarifies the defect type, but also divides the severity, making the rating results more in line with actual production needs.

[0032] S203: Analyze the change trend of the yarn real-time quality rating according to the yarn real-time quality rating to generate a subsequent quality change trend of the yarn.

[0033] The subsequent quality change trend of the yarn may refer to the direction and degree of quality change of the yarn in the subsequent textile process predicted based on the current electrostatic state, quality rating and weighted calibration results.

[0034] Specifically, the real-time quality rating of yarn can only reflect the current quality status, while textile production is a continuous dynamic process. The quality status will continue to change with the changes in many factors. If you only focus on the current quality rating, you will not be able to predict the development direction of quality in advance. This may lead to measures being taken only when the quality has seriously declined, resulting in the production of a large number of unqualified products. This step uses the real-time quality rating of yarn and then uses the known electrostatic adsorption abnormal trajectory to predict the subsequent yarn quality change trend, weaken the interference to low-sensitive categories, and generate the quality change trend on this basis. Executing targeted process control instructions based on subsequent quality change trends can timely optimize process parameters, ensure the stability of product quality, reduce the production of unqualified products, provide a basis for forward-looking process control, and provide production personnel with forward-looking risk warnings to avoid the expansion of quality problems.

[0035] S204: Execute corresponding process control instructions according to the subsequent quality change trend of the yarn, and output a quality rating log.

[0036] Process control instructions may refer to instructions for adjusting textile equipment parameters based on subsequent yarn quality trends. Quality rating logs may refer to log files that record real-time yarn quality rating results, quality trends, process control instructions, and other information.

[0037] Specifically, the quality control of traditional textile production is highly lagging, and processes are often adjusted only after defects are discovered. By this time, a certain number of defective products have already been produced, resulting in waste of raw materials and loss of efficiency. In addition, quality data is stored in a decentralized manner and lacks systematic logs, making it difficult to trace problems and achieve continuous process optimization. This step executes process control instructions in advance based on the predicted quality change trends, controlling quality problems in the bud and achieving "preemptive prevention." At the same time, the quality rating log fully records real-time rating results, trend forecasts, and control instructions and outputs them to the user, forming a closed loop of "data collection-analysis-control-recording." This not only facilitates tracing the root causes of quality problems in specific batches, but also provides long-term data support for process parameter optimization, upgrading quality control from passive response to active optimization, and significantly improving production efficiency and quality stability.

[0038] This solution first collects spatiotemporal charge data, then generates a yarn electrostatic state dataset through multimodal sensor fusion, achieving precise quantification of electrostatic factors. The electrostatic state data is then fused with morphological data to extract defect-related electrostatic representations. Real-time quality results are generated through two-dimensional dynamic rating, improving the comprehensiveness and accuracy of defect identification. Combining yarn category information and electrostatic association weights, the electrostatic data is calibrated and quality change trends are predicted, enhancing the pertinence of the analysis. Proactive process control is performed based on trends, and a quality rating log is output to the user, forming a closed-loop control system. By generating an electrostatic state dataset through multimodal sensor fusion, identifying defects and performing two-dimensional ratings based on morphological data, and then predicting and regulating trends based on category information and electrostatic weights, the dynamic details of high-speed yarns can be accurately detected and quantitatively rated in real time, enabling timely intervention in quality fluctuations, reducing the continuous output of defective products, and avoiding increased waste of raw materials, ultimately enhancing product competitiveness and reducing production costs.

[0039] In some embodiments, a spatially deployed annular array of electrostatic sensors is used to collect spatiotemporal charge datasets at key points of the yarn in real time and contactlessly; the spatiotemporal charge dataset includes real-time charge intensity values ​​and real-time charge distribution characteristics; a high-frame-rate visual sensor is used to capture yarn morphology data and motion trajectories in real time; the spatiotemporal correlation between the real-time charge intensity values, real-time charge distribution characteristics and yarn morphology data is synchronously analyzed to identify abnormal electrostatic adsorption trajectories caused by quality defects; and a yarn electrostatic state dataset is generated based on the spatiotemporal charge dataset, yarn morphology data and abnormal electrostatic adsorption trajectories.

[0040] An electrostatic sensor ring array, consisting of multiple electrostatic induction rings spaced at predetermined intervals, is used to non-contactly collect charge information on the yarn surface. The real-time charge intensity value can be the charge intensity at a specific location on the yarn at a given moment, reflecting the degree of surface charge on the yarn.

[0041] The real-time charge distribution characteristics can be the spatial distribution pattern of charge on the yarn surface (such as uniform distribution, local aggregation, etc.), reflecting the difference in charge distribution in different areas of the yarn.

[0042] A high frame rate visual sensor can be an image acquisition device with a high shooting frame rate (such as hundreds of frames per second or more), which can capture subtle morphological changes and movement trajectories during the rapid movement of yarn.

[0043] Yarn morphology data can be yarn appearance feature data collected by a high frame rate visual sensor, including yarn diameter, number of hairiness, presence of knots or breaks, and other morphological information.

[0044] Abnormal electrostatic adsorption can be caused by yarn quality defects (such as excessive hairiness or uneven fiber distribution) resulting from electrostatic adsorption, which deviates from the normal trajectory. For example, excess fibers at a defective location may produce unexpected oscillations or sticking due to electrostatic adsorption.

[0045] The yarn electrostatic state dataset can be a comprehensive dataset formed by fusing the spatiotemporal charge dataset, yarn morphology data, and electrostatic adsorption abnormal trajectory, which is used to comprehensively reflect the state of the yarn under the correlation of electrostatic properties and morphological characteristics.

[0046] Specifically, in the yarn weaving process, a single sensing method is difficult to fully reflect the nature of quality defects: if only relying on electrostatic sensing, although it can capture abnormalities in yarn surface charge (such as uneven charge distribution caused by fiber friction), it cannot distinguish whether the abnormality is caused by defects in the fiber itself (such as deviation in blending ratio) or the external environment (such as humidity changes), which is prone to misjudgment; if only relying on visual sensing, it can detect macroscopic morphological defects of yarn (such as broken ends and scars), but is not sensitive to microscopic defects (such as excessive internal hairiness), and is greatly affected by changes in light and yarn color; yarn quality defects often manifest as physical morphological abnormalities and unbalanced charge distribution (such as loose fibers accumulate excessive charges due to friction, resulting in the adsorption of impurities and causing trajectory deviation). If there is a lack of spatiotemporal correlation analysis, key diagnostic information will be missed. To address the above issues, this step first uses a spatially deployed annular array of electrostatic sensors to non-contactly sense the surface charge of the yarn through annular electrodes, converts it into a voltage signal through a charge amplifier, and then digitally generates a spatiotemporal charge data set through an ADC module; at the same time, a high-frame-rate visual sensor is used to capture yarn motion images under the illumination of a polarized light source, and the FPGA executes the hairiness segmentation algorithm in real time to output yarn morphology data (such as a diameter of 0.12mm). Based on the displacement changes of yarn feature points in continuous multi-frame images, the yarn motion trajectory (such as linear speed > 2mm / s) is generated through a three-dimensional motion reconstruction algorithm; then, the central controller receives these data streams and generates the yarn motion data according to the precision clock coordination. It is recommended to align timestamps and slice them according to time windows (e.g., 10ms) for spatiotemporal correlation analysis. When a sudden change in charge intensity is detected (e.g., ΔQ / Δt>5pC / ms), the visual system is triggered to focus on the corresponding position and the angular similarity between the charge gradient vector direction and the yarn movement offset direction is calculated. If the angular difference is small (e.g., <10°) and the offset acceleration is large (e.g., >0.5m / s²), it is determined to be an abnormal trajectory due to electrostatic adsorption. Finally, a yarn electrostatic state dataset is generated based on the analysis results. This dataset contains charge distribution characteristics, yarn morphology data, and abnormal trajectory information, and is uploaded to the manufacturing execution system through a dynamic update mechanism (e.g., generating a record every 100ms).

[0047] Through this solution, a multimodal sensing fusion strategy is adopted to enable electrostatic sensing to capture microscopic charge changes and visual sensing to record macroscopic morphological characteristics. The combination of the two can cover full-dimensional quality information from fiber-to-fiber interactions (microscopic) to the overall structure of the yarn (macroscopic); the abnormal trajectory of electrostatic adsorption associates charge changes with the motion state, which can not only identify defects but also trace their causes (for example, trajectory deviation accompanied by charge accumulation may be due to local unevenness of the yarn), providing an accurate basis for subsequent process control.

[0048] In some embodiments, based on the yarn electrostatic state data set and the real-time collected yarn morphology data, the yarn defect type is identified, and the severity of the dominant defect corresponding to the yarn defect type is determined; the yarn defect types include excessive hairiness, uneven yarn and fiber knotting; several yarn defect types are normalized to a unified numerical range to determine the normalized yarn defect type; according to the normalized yarn defect type and the severity of the dominant defect, a dynamic rating score is determined; according to the dynamic rating score, a real-time yarn quality rating is generated, and the real-time yarn quality rating is divided into three levels: excellent, medium and poor.

[0049] Yarn defect types refer to three typical quality anomalies that occur during yarn production: excessive hairiness (fibers protruding from the yarn surface), uneven yarn evenness (excessive yarn diameter fluctuation), and fiber knotting (foreign matter or fiber entanglement). The severity of a visible defect refers to the perceived severity of a particular yarn defect type under the current production conditions, typically described using quantitative indicators. Yarn real-time quality ratings refer to the yarn quality grade, classified according to dynamic rating scores, into three levels: "Excellent," "Medium," and "Poor."

[0050] Specifically, traditional yarn quality inspection often relies on a single visual feature (e.g., observing yarn shape solely through a camera), which can easily overlook electrostatic anomalies caused by changes in fiber material and friction state (e.g., excessive hairiness can lead to localized charge accumulation, while uneven yarn evenness can cause charge distribution fluctuations). Different types of yarn defects have significantly different impacts on final product performance: for example, excessive hairiness primarily affects fabric appearance, uneven yarn evenness directly reduces yarn strength, and fiber knots can cause yarn breakage during weaving. Identifying only the defect type without distinguishing its severity cannot provide an accurate basis for process adjustments (e.g., minor hairiness can be ignored, while severe hairiness requires immediate adjustment of drafting parameters). The original evaluation indicators for yarn defects have significant dimensional differences: the evaluation indicator for excessive hairiness is "number of hairs per unit length" (integer), uneven yarn evenness is "coefficient of diameter variation" (percentage), and fiber knots is "knot density" (knots / meter). A comprehensive evaluation based directly on these original indicators can lead to distorted results due to differences in numerical ranges and physical meanings (e.g., "10 hairs / meter" and "5% diameter variation" cannot be directly compared). To address the above issues, this step first uses a high-frame-rate visual sensor (e.g., 2000fps) to capture the yarn surface morphology in real time, and simultaneously combines it with the charge distribution heat map generated by the electrostatic sensor ring array to accurately locate three types of defect areas: excessive hairiness, uneven yarn length, and fiber knots. Subsequently, the number of hairs per unit length (e.g., 15 pieces / cm) and the average charge intensity (e.g., 0.5nC) are calculated for the hairiness area, the diameter fluctuation variance (e.g., 0.2mm²) is extracted for the uneven yarn length section, and the charge peak value (e.g., 8nC) and the geometric size of the impurities (e.g., 0.3mm²) are recorded for the fiber knot points. Next, the historical extreme value library of the current production batch (e.g., The hairiness density extremes range from 0 to 85 strands / cm). The original defect values ​​are converted to a uniform interval (e.g., [0, 1]) using the range normalization formula. Based on the real-time defect severity (e.g., a coefficient of 0.5 for moderate hairiness defects), the normalized values ​​are multiplied by the corresponding weight (e.g., a hairiness weight of 0.3) and the severity coefficient, and the sum is accumulated (e.g., 0.18 × 0.3 × 0.5 ≈ 0.027) to obtain the total dynamic rating score (e.g., 0.45). Finally, the rating result is mapped to a two-dimensional rating: an excellent rating is assigned for a score ≤ 0.3, an intermediate rating is assigned for a score between 0.3 and 0.7 (e.g., 0.45), and a poor rating is assigned for a score greater than 0.7.

[0051] Through this solution, by integrating electrostatic state data and morphological data, not only can the physical morphological defects of yarn (such as hairiness and yarn evenness) be identified, but also hidden defects (such as charge anomalies caused by fiber knots) can be captured through changes in electrostatic characteristics, avoiding the limitations of a single data source and improving the comprehensiveness and accuracy of quality defect identification; dynamic rating scores are updated in real time with the production process, which can promptly reflect quality fluctuations and are more in line with actual production status than traditional static sampling ratings; normalization processing eliminates the evaluation differences of different defects, and weighted calculation enables the rating results to better reflect the comprehensive impact of defects on quality, providing an accurate basis for process adjustments.

[0052] In some embodiments, the yarn surface image and the corresponding real-time charge intensity value are obtained in real time, and the following hairiness density grid analysis steps are performed based on the yarn surface image: when the number of discrete fibers in several consecutive detection data exceeds a preset first density threshold and the average electrostatic intensity thereof is greater than a preset multiple of the number of visual hairiness, it is determined to be a local hairiness aggregation defect; when the number of discrete fibers in a unit detection area exceeding a preset proportion exceeds a preset second density threshold and the overall electrostatic distribution unevenness is greater than a preset diffuse hairiness ratio, it is determined to be a diffuse hairiness excess defect; based on the combination pattern of the number of local aggregation areas and the diffuse distribution ratio, combined with the coupling coefficient of the electrostatic intensity increase and the hairiness density, the spatial distribution state of the hairiness is analyzed, and the degree of hairiness density dominance is quantified.

[0053] The yarn surface image can be an image of the yarn surface taken by a high-frame-rate visual sensor, which can intuitively present the appearance characteristics of the yarn, such as hairiness distribution and fiber arrangement, and is a direct basis for visual analysis of hairiness defects.

[0054] Real-time charge intensity data, collected at key points on the yarn by a circular array of spatially deployed electrostatic sensors, reflects the strength of static electricity generated by hairiness friction and other factors, and is strongly correlated with hairiness density. The preset first density threshold can be a pre-set critical value for the number of discrete fibers used to determine whether hairiness is "aggregated" in a local area. This threshold, derived from historical high-quality yarn data, ensures a consistent standard for determining localized aggregation defects.

[0055] A localized hairiness accumulation defect can refer to a sudden increase in the number of discrete fibers and significantly elevated static electricity within a localized area on the yarn surface. This defect is typically caused by uneven combing of the fibers, significantly impacting the risk of "thread jamming" during subsequent weaving. The preset second density threshold can be a pre-set critical value for the number of discrete fibers used to determine whether the hairiness within a unit detection area is "excessive." This value is typically lower than the preset first density threshold (because the hairiness count in a single area of ​​a diffuse defect is lower than that of a localized accumulation, but is distributed over a wider area). A diffuse excess hairiness defect can refer to an excessive number of discrete fibers in multiple areas of the yarn surface (exceeding a preset ratio) and a chaotic overall static electricity distribution. This defect is typically caused by poor fiber raw material quality or an unstable drafting process, significantly impacting yarn strength and dyeing uniformity.

[0056] The spatial distribution state of hairiness can refer to the spatial characteristics of the distribution position, range and density of hairiness on the yarn surface, such as "local concentration" and "global dispersion", which is the key basis for distinguishing defect types.

[0057] The degree of hairiness density visibility can be comprehensively characterized by quantitative indicators (such as the number of local aggregation areas, diffuse distribution ratio, coupling coefficient, etc.) to measure the obviousness of hairiness defects in appearance and electrostatic characteristics. It is the core basis for determining the severity of dominant defects.

[0058] Specifically, hairiness is one of the core indicators of yarn quality. Its quantity and distribution directly affect the stability of subsequent weaving processes (such as breakage rate and fabric flatness) and dyeing (such as dyeing levelness), as well as the appearance and performance of the final product. Traditional yarn hairiness detection methods mainly rely on visual image analysis, which has two major limitations: first, they are easily affected by light and image noise, leading to "misjudgments" (such as misidentifying dust as hairiness); second, they judge the severity of hairiness only by the number of hairiness, ignoring the distribution characteristics of hairiness (localized accumulation and global dispersion have significant differences in their impact on the process), resulting in a lack of targeted subsequent process control. The causes and impacts of localized hairiness accumulation (such as the sudden appearance of a large amount of hairiness in a 1mm area) and excessive diffuse hairiness (such as a small amount of hairiness throughout the yarn but the overall amount exceeds the standard) are completely different. If these are not distinguished, and the same process control strategy is adopted uniformly, it will only lead to "treating the symptoms but not the root cause." To address the above issues, the implementation process of this step first performs defect determination: by grid segmenting the yarn surface image, the number of discrete fibers in each unit detection area (such as 1mm×1mm) is counted; for local hairiness aggregation defects, when the number of fibers in a certain area exceeds the preset first density threshold (such as 50 fibers / mm²) for several consecutive detections (such as 5 consecutive times) and the average electrostatic intensity of the area is greater than the preset multiple of the visual hairiness number (such as 1.2 times), it is marked as an aggregation defect area and the number is recorded; for diffuse hairiness excess defects, the number of fibers in all unit areas exceeding the preset second density threshold is calculated. If the excess ratio exceeds a preset ratio (e.g., 20%) and the overall electrostatic distribution unevenness is greater than a preset diffuse hairiness ratio (e.g., 0.4), it is determined to be a diffuse defect. The hairiness spatial distribution is then analyzed: classification is performed based on the combination pattern of the number of local clustered areas and the diffuse distribution ratio (e.g., pattern A is high clustered number + low diffuse ratio, and pattern B is the opposite). The degree of dominance is quantified based on the coupling coefficient (e.g., electrostatic intensity increase / hairiness density increase). The specific formula is: Dominance = (number of clustered areas × 0.6 + diffuse ratio × 0.4) × coupling coefficient.

[0059] This solution, utilizing the dual-dimensional analysis of "visual image + electrostatic characteristics," effectively improves the accuracy of hairiness defect identification and reduces misjudgments due to other external factors. It also achieves refined classification of hairiness defects, accurately distinguishing between localized hairiness accumulation and diffuse hairiness excess, providing a clear direction for subsequent process control and avoiding blind adjustments. By comprehensively considering indicators such as the number of localized accumulation areas, the coupling coefficient between the electrostatic intensity increase and hairiness density, the quantification of the severity of dominant defects is made more scientific, providing accurate data support for quality rating, thereby enhancing the controllability of the production process, helping to reduce problems such as weaving breakage and uneven dyeing caused by hairiness defects, and improving the stability and consistency of yarn quality.

[0060] In some embodiments, a yarn axial image sequence is acquired in real time, and the following yarn three-dimensional fluctuation analysis steps are performed based on the yarn axial image sequence: a diameter abnormal area where the diameter difference between adjacent sampling points exceeds the diameter tolerance threshold is marked as a sudden uneven area, and the angular change rate of the electrostatic adsorption trajectory of the abnormal diameter area is synchronously detected. When the angular change rate exceeds the preset turbulence threshold, the current diameter abnormal area is marked as an electrostatic turbulence type sudden uneven area; a trend abnormal area corresponding to a unidirectional continuous increase / decrease diameter trend exceeding a preset number of points is marked as a gradual uneven area, and the electrostatic gradient distribution slope of the gradual area is synchronously acquired. When the slope is in the same direction as the diameter change trend and exceeds the associated threshold, it is marked as an electrostatic cumulative type gradual uneven area; the proportion of the length of the sudden uneven area and the length of the gradual uneven area in the unit detection area is determined, and the degree of the dominance of the yarn unevenness is quantified by combining the maximum diameter deviation value and the electrostatic turbulence intensity parameter which are positively correlated with the dominance of the yarn unevenness, and the electrostatic gradient consistency parameter which is negatively correlated with the dominance of the yarn unevenness.

[0061] A yarn axial image sequence can be a collection of axial cross-sectional images of the yarn surface captured continuously along the direction of yarn motion. These images are collected at fixed intervals using a high-frame-rate visual sensor, forming time-series image data reflecting the continuous changes in yarn diameter. Three-dimensional yarn fluctuation analysis can be a method that analyzes the fluctuation characteristics of yarn diameter variations in the three-dimensional spatial dimension, combined with changes in electrostatic parameters, to identify the type and distribution of yarn unevenness. The diameter tolerance threshold can be a preset maximum allowable difference in yarn diameter variation between adjacent sampling points. Exceeding this value is considered a diameter anomaly.

[0062] The sudden uneven area may refer to a diameter abnormal area where the diameter difference between adjacent sampling points in the yarn axis exceeds the diameter tolerance threshold, which is manifested as a significant sudden change in diameter within a short distance.

[0063] The rate of change of the direction angle of the electrostatic adsorption track may refer to the degree of change of the direction angle of the track formed by the electrostatic adsorption of the surrounding particles on the yarn surface within a unit time, reflecting the stability of the electrostatic field.

[0064] Electrostatic turbulence-type mutation unevenness can refer to a defect type in which the rate of change of the electrostatic adsorption trajectory direction angle exceeds a preset turbulence threshold in the mutation-type uneven area, indicating that diameter mutation is strongly correlated with electrostatic field disorder.

[0065] The gradual uneven area may refer to an abnormal trend area corresponding to a unidirectional continuous increase / decrease trend in the yarn axial direction exceeding a preset number of points, which is manifested as a slow change in diameter over a long distance.

[0066] Static accumulation-type gradual unevenness refers to a defect where the slope of the static gradient distribution and the diameter change trend within the gradual unevenness area are in the same direction and exceed the correlation threshold, indicating a strong correlation between diameter gradualness and static accumulation / dissipation. Maximum diameter deviation refers to the maximum difference between the yarn diameter and the standard diameter within a unit test area, directly reflecting the severity of the yarn unevenness.

[0067] The electrostatic disorder intensity parameter can refer to a quantitative indicator used to characterize the degree of disorder of the electrostatic field in a unit detection area, which integrates characteristics such as the rate of change of the angular direction of the electrostatic adsorption trajectory and the electrostatic gradient fluctuation.

[0068] The electrostatic gradient consistency parameter may refer to a quantitative indicator used to characterize the uniformity of the electrostatic gradient distribution within a unit detection area. The higher the consistency, the more stable the electrostatic field.

[0069] The degree of visibility of yarn unevenness can refer to the degree of manifestation of yarn unevenness defects in yarn morphology and electrostatic characteristics, and is an indicator of defect severity quantified by comprehensive multiple parameters.

[0070] Specifically, the traditional method only distinguishes between sudden and gradual defects by the characteristics of diameter change. However, in actual production, some diameter sudden changes may be accidental fluctuations caused by short-term mechanical vibrations, rather than real quality defects. Some gradual trends may cause "false diameter changes" due to electrostatic adsorption of fibers. If only relying on morphological data, normal fluctuations may be misjudged as defects, or hidden defects related to static electricity may be missed, resulting in a high misjudgment rate of defect types. Moreover, the traditional method only measures the severity of uneven yarn length by diameter deviation value, but ignores the amplification effect of electrostatic characteristics on defects. For example, under the same diameter deviation, if accompanied by strong electrostatic disorder, it may cause further aggregation of fibers in the subsequent weaving process, and the defect will continue to worsen. If the electrostatic field is stable, the defect may remain stable. Relying only on the quantitative results of diameter parameters cannot reflect the "potential deterioration risk" of defects, resulting in insufficient accuracy in quality rating. To address the above issues, this step first uses a high-frame-rate visual sensor to capture an axial image sequence at a fixed frame rate (e.g., 500 frames / second) along the direction of yarn travel, and simultaneously uses a circular array of electrostatic sensors to synchronously collect the electrostatic adsorption trajectory data of the corresponding section; then, the yarn diameter is extracted frame by frame from the axial image sequence and the diameter difference of adjacent sampling points is calculated. When the difference exceeds a preset diameter tolerance threshold (e.g., ±8% of the nominal diameter), it is marked as a candidate mutation area, and the angular change rate of the electrostatic adsorption trajectory in the area is simultaneously analyzed. If the change rate exceeds a preset turbulence threshold (e.g., ≥15° / millisecond), it is determined to be an electrostatic turbulence-type mutation unevenness, otherwise it is marked as a normal mutation area; at the same time, the image is detected. The areas with a unidirectional diameter continuous increase or decrease trend that exceed the preset number of points (such as ≥5 sampling points) in the sequence are marked as candidate gradual change areas and their electrostatic gradient distribution slope is calculated. When the slope is in the same direction as the diameter change trend (such as the slope is >0 when the diameter increases) and the absolute value exceeds the associated threshold (such as ≥0.3), it is determined to be electrostatic cumulative type gradual unevenness; finally, the proportion of sudden change type and gradual change type areas within the unit detection length (such as 1 meter) is statistically analyzed, and positive correlation parameters such as maximum diameter deviation value, electrostatic turbulence intensity (such as the mean value of the directional angle change rate) and negative correlation parameters such as electrostatic gradient consistency (such as the inverse of the slope variance of adjacent areas) are integrated to finally quantify the degree of strip unevenness between 0 and 1.

[0071] This solution integrates yarn morphology data and electrostatic characteristics to accurately distinguish between sudden and gradual yarn unevenness, identify whether the defect is dominated by static electricity, significantly reduce the misjudgment rate, and improve the fit between defect type identification and actual production causes. It also quantifies the degree of visibility of yarn unevenness by integrating multi-dimensional parameters, reflecting both the current severity and the potential risk of deterioration, making the results more comprehensive and predictive. Once the type and cause of the yarn unevenness defect are clarified, subsequent process control can be adjusted in a targeted manner to improve quality improvement efficiency.

[0072] In some embodiments, yarn surface scanning imaging data is acquired in real time, and the corresponding electrostatic intensity extreme value is captured synchronously, and the following impurity multi-scale feature extraction steps are performed to clarify the impurity attributes: when there is a dark clumping structure with irregular boundaries whose charge decay rate is less than a preset decay rate threshold, the dark clumping structure is determined to be a fiber agglomerate; when there are reflective particles with geometric edges whose electrostatic intensity extreme value is greater than a preset multiple of the particle size, the reflective particles are determined to be hard impurities; when there is a radial fiber entanglement structure whose electrostatic intensity is in an oscillating and fluctuating state, the radial fiber entanglement structure is determined to be a yarn defect nodule; based on the size grade, optical contrast and boundary clarity of the impurity multi-scale characteristics, the size grade, optical contrast and boundary clarity are weighted according to the impurity attributes to quantify the degree of fiber knot visibility.

[0073] Yarn surface scanning imaging data may refer to image sequence data containing surface details (such as texture, foreign matter, structural anomalies, etc.) obtained by continuously scanning the yarn surface through a high frame rate visual sensor.

[0074] The extreme value of electrostatic intensity may refer to the maximum value or minimum value (ie, the extreme value of electrostatic intensity) among the electrostatic intensity values ​​at different positions on the yarn surface collected synchronously during the yarn surface scanning process.

[0075] The charge decay rate can be measured as the time required for the static charge intensity on the yarn surface to decrease from its peak value to a stable value. Fiber agglomeration refers to the irregular clumps of fibers that form during yarn production due to entanglement and adhesion. These clumps are made of the same material as the yarn itself (mostly fibers) and exhibit strong fiber aggregation characteristics.

[0076] Hard impurities can refer to non-fibrous hard foreign matter (such as metal debris, plastic particles, sand, etc.) mixed into the yarn, which has clear geometric edges and higher hardness.

[0077] Yarn defect nodules can refer to the radial entangled structure formed by abnormal fiber arrangement during the twisting or stretching process of the yarn. It manifests as local loose fibers and entangled knots, which is a structural defect of the yarn itself.

[0078] Size grade refers to a grading standard based on the physical dimensions of an impurity (e.g., length, diameter, area, etc.), used to characterize the degree to which the physical size of an impurity affects yarn quality. Optical contrast refers to the difference in optical signals (e.g., grayscale value, brightness, color saturation, etc.) between the impurity area and the yarn proper in the scanned imaging data. The greater the difference, the easier the impurity is to visually identify. Boundary clarity refers to the clarity of the boundary between the impurity and the yarn proper. This is determined by image edge detection algorithms, which calculate parameters such as boundary continuity and sharpness. The sharper the boundary, the more distinguishable the impurity from the yarn proper.

[0079] The degree of fiber contamination visibility refers to the degree to which impurities appear on the yarn surface, and is a quantitative representation of the impact of impurities on yarn quality (the higher the value, the more significant the negative impact of impurities on yarn quality).

[0080] Specifically, in yarn textile quality rating, impurities are one of the key factors affecting yarn performance (such as strength, uniformity, and subsequent weaving stability), but the impact mechanisms of different types of impurities on quality are significantly different: fiber agglomeration can lead to local uneven thickness of the yarn, which can easily cause breakage or fabric pilling during the weaving process; hard impurities may wear textile equipment (such as the reed and rollers of the loom), shortening the life of the equipment, and may also form holes on the fabric surface; yarn nodules will destroy the continuity of the yarn, resulting in uneven dyeing or bumps on the fabric surface, affecting the appearance quality. If "impurities" are only used as a quality defect indicator in general without distinguishing their properties and quantifying their impact, the quality rating results will be distorted. For example, fiber agglomerates and hard impurities of the same size, the latter are far more harmful to production than the former. If they are treated equally, the process control instructions will lose their pertinence (for example, hard impurities require emergency shutdown and cleaning, while fiber agglomerations can be alleviated by adjusting the drafting parameters). To address the above issues, this step synchronously triggers a high-frame-rate visual sensor and an electrostatic sensor ring array to collect yarn surface scanning imaging data and the corresponding electrostatic intensity extremes at a high frequency (e.g., 2kHz) in real time. Multi-scale feature extraction of impurities is then performed to clarify their attributes: when a dark, clumping structure with irregular boundaries is located in the imaging data, and the charge decay rate of its corresponding electrostatic signal is less than a preset threshold (e.g., 50V / s) for several frames (e.g., 3 frames), it is determined to be a fiber agglomeration; when a reflective particle area with geometric edges is detected, and the ratio of its electrostatic intensity extreme to its physical size is greater than a preset multiple (e.g., 8 times for metal impurities or 5 times for gravel impurities), it is determined to be a hard impurity; when a radial fiber entanglement structure is identified, and its corresponding electrostatic signal decays within a short time (e.g., 0. When a certain number of oscillation fluctuations (amplitude change rate such as >20%) appear within 5 seconds (e.g., ≥3 times), it is determined to be a yarn defect knot; then, according to the identified impurity attributes, its size grade (e.g., micro=0.1, small=0.3, medium=0.6, large=1.0), optical contrast (normalized value) and boundary clarity (normalized gradient value) are extracted respectively, and weighted summation is performed according to the preset weights corresponding to the attributes (e.g., agglomeration: size weight such as 0.5, contrast weight such as 0.3, boundary weight such as 0.2; hard impurities: size weight such as 0.4, contrast weight such as 0.4, boundary weight such as 0.2; knots: size weight such as 0.6, contrast weight such as 0.2, boundary weight such as 0.2), and finally a quantified fiber knot visibility value ranging from 0 to 1 is obtained.

[0081] Through this solution, the yarn surface scanning imaging data and the extreme values ​​of electrostatic intensity are integrated, and combined with multi-scale feature extraction, it is possible to effectively distinguish between fiber agglomerates, hard impurities, and yarn defect nodules, avoid missed detection or misjudgment due to single visual detection, and improve the accuracy of impurity identification; by weighting the size grade, optical contrast, and boundary clarity, the degree of influence of different impurities on yarn quality can be scientifically quantified, so that the quality rating results are more in line with actual production needs, and the accurate quantification of quality impact is achieved; after clarifying the impurity attributes and degree of visibility, process control instructions can be generated in a targeted manner, supporting the precise control of the process, thereby reducing ineffective shutdowns or excessive control caused by impurity misjudgment, reducing production costs, and at the same time reducing the wear of equipment by hard impurities, extending the service life of textile equipment, and improving production efficiency.

[0082] In some embodiments, based on a preset rule table, several identified yarn defect types are mapped to defect type identifiers, and the assessed severity of dominant defects is mapped to severity level identifiers; several defect type identifiers are string-concatenated with corresponding severity level identifiers, and the concatenated result is used as a unit rating code; after all unit rating codes are concatenated, several unit rating codes are string-concatenated a second time to generate a composite rating code; according to preset mapping rules, the composite rating code is mapped to an initial quality grade of excellent / medium / poor; the mapping rules include: determining the total number of all defects as the defect density, and identifying the highest severity level identifier in the composite rating code; if the defect density is greater than a first density threshold or the highest severity level identifier reaches the first severity level threshold, it is judged as poor; if the defect density does not meet the aforementioned condition for being judged as poor but is greater than a second density threshold or the highest severity level identifier reaches the second severity level threshold, it is judged as medium; if the defect density does not exceed the second density threshold and the highest severity level identifier does not reach the second severity threshold, it is judged as excellent; wherein, the first density threshold is greater than the second density threshold, and the first severity level threshold is higher than the second severity threshold.

[0083] The preset rule table can be a pre-established mapping relationship table for converting yarn defect types and severities into standardized identifiers. The defect type identifier can be a unique code representing the yarn defect category (e.g., "B" for excessive hairiness, "C" for uneven yarn evenness). The severity level identifier can be a hierarchical code representing the severity of the defect (e.g., "1" for mild, "2" for severe).

[0084] The unit rating code can be the complete rating identification of a single defect, which is composed of the defect type identification and the severity level identification (such as "C1" for slight unevenness).

[0085] The composite rating code can be a global identifier that integrates all individual rating codes (e.g., "B3+C1" indicates severe excessive hairiness and slight uneven yarn evenness). The first density threshold can be the critical defect count threshold for determining yarn quality as "poor," representing a preset higher defect count standard. The second density threshold can be the critical defect count threshold for determining yarn quality as "medium," and is lower than the first density threshold, representing a preset medium defect count standard. The first severity threshold can be the severity threshold for determining yarn quality as "poor," representing a preset higher severity level standard. The second severity threshold can be the severity threshold for determining yarn quality as "medium," and is lower than the first severity threshold.

[0086] Specifically, quality rating is the core link between quality inspection and process control, which directly affects production efficiency, product qualification rate and production cost. Traditional yarn quality rating methods have limitations: first, there is the one-sidedness of single defect evaluation. Traditional methods often rate single defects (such as only evaluating the number of hairiness), ignoring the comprehensive impact of multiple defects on the overall quality. For example, a yarn may have both slight excessive hairiness and moderate uneven yarn length. The single defect rating cannot reflect its true quality level; second, the ambiguity of grade division and the lack of clear "excellent, medium, and poor" grade judgment thresholds make it impossible for production personnel to accurately judge whether the yarn meets production requirements, thereby delaying the opportunity for process adjustment and increasing the defective rate. To address the above issues, this step first calls the preset rule table to map the identified defect type to a standardized identifier (e.g., excessive hairiness is mapped to "B"), and generates a severity level identifier (e.g., "3") based on the severity of the dominant defect (e.g., the hairiness density value reaches level 3). A string concatenation operation is then performed on each defect to generate a unit rating code (e.g., "B3") containing type and level information. All defects are then traversed and the unit rating codes are concatenated again according to the detection sequence to form a global composite rating code (e.g., "B2+C1", indicating two types of defects: excessive hairiness (level 2) and uneven yarn evenness (level 1). The composite rating code is then statistically analyzed. The total number of defect type identifiers in the rating code is used as the defect density (for example, 2 defects / meter), and all severity level identifiers are parsed to extract the highest value (in this case, "2"). Finally, the quality grade is determined based on a dual-threshold mechanism: if the defect density exceeds the first density threshold (for example, 5 defects / meter) or the highest severity level reaches the first severity level threshold (for example, "3"), the product is judged as "poor". If it does not reach "poor" but the defect density exceeds the second density threshold (for example, 2 defects / meter) or the highest severity level reaches the second severity level threshold (for example, "2"), the product is judged as "fair". Only when the defect density is ≤2 defects / meter and the highest severity level is lower than "2" is the product judged as "excellent".

[0087] Through this solution, all defect information is integrated using composite rating codes, avoiding the limitations of single defect assessment and truly reflecting the overall quality status of the yarn. The quality grade boundaries are clearly defined, making it easier for production personnel to quickly identify the yarn quality status and take corresponding measures in a timely manner, reducing production losses caused by ambiguous quality judgments. Real-time process control is supported, providing a clear basis for process optimization, helping to reduce the defective rate and improve the quality stability and production efficiency of yarn products. At the same time, standardized management of quality data is promoted. Standardized coding methods such as composite rating codes facilitate the storage, transmission and analysis of quality data, laying the foundation for subsequent quality traceability, production optimization and big data analysis.

[0088] In some embodiments, based on the abnormal trajectory of electrostatic adsorption, the trajectory time series dynamic characteristics including trajectory fluctuation frequency, directional consistency and intensity variation amplitude are extracted; the trajectory evolution mode is identified based on the dynamic characteristics: when the trajectory fluctuation frequency is lower than the preset low-frequency threshold and the directional consistency is higher than the stability threshold, it is a steady slowdown mode; when the trajectory fluctuation frequency is higher than the preset high-frequency threshold and the intensity variation amplitude continues to increase, it is a deterioration diffusion mode; when the directional consistency is lower than the disorder threshold and the intensity variation amplitude periodically decreases, it is an intermittent oscillation mode; the trajectory evolution mode is mapped to a quality trend: the steady slowdown mode is mapped to a defect alleviation trend, the deterioration diffusion mode is mapped to a defect escalation trend, and the intermittent oscillation mode is mapped to a quality fluctuation trend; according to the quality trend combined with the severity of the dominant defect, the subsequent quality change trend of the yarn is generated.

[0089] An abnormal electrostatic adsorption trajectory can refer to the path of an electrostatic adsorption trajectory caused by a yarn quality defect, where the trajectory deviates from its normal state. Trajectory temporal dynamic features can refer to a set of features that describe the changing patterns of an abnormal electrostatic adsorption trajectory over a time series. Trajectory fluctuation frequency can refer to the number of times the abnormal electrostatic adsorption trajectory fluctuates per unit time, reflecting the frequency of trajectory changes. Directional consistency can refer to the degree to which the abnormal electrostatic adsorption trajectory maintains a consistent direction during its movement; higher values ​​indicate a more stable trajectory direction. Intensity variation can refer to the range of variation in the electrostatic intensity corresponding to the abnormal electrostatic adsorption trajectory over a time series, reflecting the magnitude of the fluctuation in electrostatic intensity. Trajectory evolution pattern can refer to the typical temporal patterns of abnormal electrostatic adsorption trajectory changes, as summarized based on the trajectory temporal dynamic features. A steady-slowing pattern can refer to an evolutionary state characterized by low trajectory fluctuation frequency and high directional consistency. A deteriorating diffusion pattern can refer to an evolutionary state characterized by high trajectory fluctuation frequency and a continuously increasing intensity variation. An intermittent oscillation pattern can refer to an evolutionary state characterized by low trajectory directional consistency and a periodic decrease in intensity variation. A defect alleviation trend can refer to a gradual reduction in yarn quality defects. A defect escalation trend can refer to a gradual increase in yarn quality defects. The quality fluctuation trend can be a trend in which yarn quality defects change repeatedly and are unstable.

[0090] Specifically, real-time quality rating can only reflect the quality status at a certain moment, while yarn quality is a dynamic process: factors such as wear and tear of textile equipment, fluctuations in raw material properties, and changes in ambient temperature and humidity can cause yarn defects to evolve from "mild" to "serious" or from "stable" to "fluctuating". If process control relies solely on real-time quality rating, it can often only "passively respond" to defects that have already occurred and cannot prevent the expansion or deterioration of defects in advance, which may lead to a large number of unqualified products and increase production costs. For example, when the yarn has slight "unevenness", the real-time quality rating may be "medium", but if its electrostatic adsorption trajectory has shown the characteristics of "increased frequency of fluctuations and increased intensity variation" (indicating that the defect will escalate), if no timely intervention is made, it may develop into serious unevenness in a short period of time, and the rating will be downgraded to "poor". At this time, adjusting the process will result in a large amount of yarn waste. In response to the above problems, this step and this embodiment first obtains the electrostatic adsorption abnormal trajectory data, and then extracts key dynamic features from the time series information of the trajectory: the trajectory fluctuation frequency (such as 0.5Hz) is calculated by analyzing the trajectory direction / speed changes, the direction consistency (such as 0.85, the higher the value, the more consistent) is calculated by counting the discrete degree of the direction vector in the continuous time window, and the intensity variation amplitude (such as 0.3) is determined by calculating the range or standard deviation of the adsorption intensity signal in the window; based on these quantitative features, combined with the preset discrimination thresholds (such as the low-frequency threshold of 0.3Hz for fluctuation frequency, the high-frequency threshold of 1.2Hz, the direction consistency stability threshold of 0.8, the disorder threshold of 0.6) and the change trend of the intensity variation amplitude (continuous increase, periodic decrease or stability), the trajectory evolution mode is identified - if the fluctuation frequency is lower than the low-frequency threshold (such as 0.25Hz) and the direction consistency is higher than the stability threshold (such as 0.85), it is determined to be a steady slowing mode; if the fluctuation If the frequency is higher than the high-frequency threshold (e.g., 1.3 Hz) and the intensity variation amplitude continues to increase (e.g., the previous value increases from 0.2 to 0.35), it is determined to be a deterioration diffusion mode. If the directional consistency is lower than the disorder threshold (e.g., 0.55) and the intensity variation amplitude decreases periodically (e.g., the amplitude periodically decreases from 0.4 to 0.1), it is determined to be an intermittent oscillation mode. The identified pattern is then directly mapped to the corresponding quality trend (steady slowdown -> defect relief trend, deterioration diffusion -> defect escalation trend, intermittent oscillation -> quality fluctuation trend). Finally, combined with the current severity of the dominant defect evaluated in the aforementioned embodiment (e.g., hairiness defect severity of 0.7, unevenness defect severity of 0.4), a comprehensive subsequent yarn quality change trend is generated (e.g., "Prediction: Defect escalation trend (based on deterioration diffusion mode), current hairiness defect severity is high (0.7), requiring immediate intervention"), providing a decision-making basis for process control that includes future prediction direction and current severity.

[0091] Through this solution, the quality trend is corrected by analyzing the trajectory evolution model and combining it with the severity of the dominant defects. It is possible to predict the changing trend of yarn quality defects before they deteriorate significantly, and adjust the strategy to be more in line with the actual production situation, thereby buying time for process adjustments and reducing the production of unqualified products. By analyzing the changing trend of yarn quality and combining "real-time rating" with "future prediction", a shift from "passive response" to "active prevention" is achieved, which is irreplaceable for improving the stability of yarn textile quality, reducing production costs, and improving production efficiency.

[0092] In some embodiments, differentiated control is performed according to the subsequent quality change trend type of the yarn: if it is determined to be a defect escalation trend, linkage control is triggered to reduce the spinning speed and increase the environmental humidification; if it is determined to be a defect alleviation trend, the current process is maintained and the detection density is reduced; if it is determined to be a quality fluctuation trend, the ambient temperature and humidity are dynamically adjusted and static electricity neutralization is started in a targeted manner; the real-time quality rating of the yarn, the subsequent quality change trend of the yarn and the control instructions are associated with the production line batch and time, and a quality rating log is generated and output.

[0093] The quality change trend type may refer to the specific direction of subsequent quality changes of the yarn, including defect escalation trend, defect alleviation trend and quality fluctuation trend.

[0094] The quality rating log may refer to a document that records the real-time quality rating of the yarn, quality change trends, control instructions, production line batches and time, which is used for quality traceability and process optimization.

[0095] Specifically, in the yarn weaving process, quality control is not a single "detection-rating" closed loop, but requires forward-looking regulation through the prediction of quality change trends, and continuous optimization supported by complete log records. The lack of complete logs will lead to the inability to trace quality problems and the difficulty in accumulating control experience, which will restrict the continuous improvement of the production line. In response to the above problems, this step receives the subsequent quality change trend data of the yarn in real time, and parses its trend classification labels (upgrade / relief / fluctuation), and then matches the preset control rule library: If it is determined to be a defect escalation trend, a speed reduction instruction is sent to the spinning equipment (such as from 300m / min to 250m / min) and the humidification system is simultaneously triggered to increase the ambient humidity (such as from 50% to 65%); if it is determined to be a defect relief trend, the current process parameters are maintained (such as spinning speed 300m / min, humidity 50%) and the sampling frequency of the detection sensor is reduced frequency (e.g., from 100Hz to 50Hz); if a quality fluctuation trend is determined, the ambient temperature and humidity are dynamically adjusted (e.g., fine-tuned within the range of ±2°C and ±5%), and the ion generator is activated to spray neutralizing airflow toward the area with abnormal static electricity; all control instructions are sent to the equipment controller via the industrial bus, while the yarn static electricity status is monitored in real time to verify the effectiveness of the control; ultimately, a structured record containing batch number, timestamp, real-time rating (e.g., "excellent"), trend type (e.g., "fluctuation"), control instruction (e.g., "temperature and humidity +2°C"), and execution status is output.

[0096] Through this solution, the differentiated regulation of yarn quality change trends and the output of quality rating logs can improve the accuracy of quality control, take targeted measures for different quality change trends, avoid the drawbacks of "one-size-fits-all" regulation, quickly curb defect escalation, stabilize quality fluctuations, optimize resource allocation, and significantly improve the stability and consistency of yarn quality; at the same time, strengthen quality traceability and process optimization. The complete log provides a basis for tracing the root causes of quality problems and summarizing the best process parameter combinations, which helps to continuously improve production; and reduce the overall production cost from multiple dimensions such as reducing defective products, reducing energy consumption and equipment loss, and reducing large-scale rework.

[0097] Figure 3 This is a structural diagram of a real-time quality rating system for yarn textiles provided in one embodiment of the present application, as shown in FIG. Figure 3 As shown, a real-time quality rating system 300 for yarn weaving of this embodiment includes: a charge monitoring module 301 , a two-dimensional rating module 302 , a trend deduction module 303 and a control execution module 304 .

[0098] The charge monitoring module 301 is used to obtain a spatiotemporal charge data set and generate a yarn electrostatic state data set through a multimodal sensing fusion strategy; the two-dimensional rating module 302 is used to identify yarn textile quality defects based on the yarn electrostatic state data set and in combination with the real-time collected yarn morphology data, and perform two-dimensional dynamic rating according to the defect type and severity to generate a yarn real-time quality rating; the trend deduction module 303 is used to analyze the change trend of the yarn real-time quality rating based on the yarn real-time quality rating and generate a subsequent quality change trend of the yarn; the control execution module 304 is used to execute corresponding process control instructions according to the subsequent quality change trend of the yarn and output a quality rating log.

[0099] Optionally, the charge monitoring module 301 is specifically used to: utilize a spatially deployed annular array of electrostatic sensors to collect a spatiotemporal charge dataset on key points of the yarn in real time and non-contact; the spatiotemporal charge dataset includes a real-time charge intensity value and a real-time charge distribution characteristic; utilize a high frame rate visual sensor to capture yarn morphology data and motion trajectory in real time; synchronously analyze the spatiotemporal correlation between the real-time charge intensity value, the real-time charge distribution characteristic and the yarn morphology data to identify abnormal electrostatic adsorption trajectories caused by quality defects; and generate the yarn electrostatic state dataset based on the spatiotemporal charge dataset, the yarn morphology data and the abnormal electrostatic adsorption trajectories.

[0100] Optionally, the two-dimensional rating module 302 identifies yarn textile quality defects based on the yarn electrostatic state data set and the yarn morphology data collected in real time, and performs two-dimensional dynamic rating according to the defect type and severity to generate the yarn initial quality rating. It is specifically used to: identify the yarn defect type based on the yarn electrostatic state data set and the yarn morphology data collected in real time, and determine the severity of the dominant defect corresponding to the yarn defect type; the yarn defect type includes excessive hairiness, uneven yarn and fiber knotting; normalize several of the yarn defect types to a unified numerical range to determine the normalized yarn defect type; determine the dynamic rating score according to the normalized yarn defect type and the severity of the dominant defect; generate a yarn real-time quality rating based on the dynamic rating score, and the yarn real-time quality rating is divided into three levels: excellent, medium and poor.

[0101] Optionally, when the two-dimensional rating module 302 identifies the type of yarn defect based on the yarn electrostatic state data set and the yarn morphology data collected in real time, and determines the severity of the dominant defect corresponding to the yarn defect type, it is specifically used to: obtain the yarn surface image and the corresponding real-time charge intensity value in real time, and perform the following hairiness density grid analysis step based on the yarn surface image: when the number of discrete fibers in several consecutive detection data exceeds a preset first density threshold and the average value of their electrostatic intensity is greater than a preset multiple of the number of visual hairiness, it is determined to be a local hairiness aggregation defect; when the number of discrete fibers in a unit detection area exceeding a preset proportion exceeds a preset second density threshold and the overall electrostatic distribution unevenness is greater than a preset diffuse hairiness ratio, it is determined to be a diffuse hairiness excess defect; according to the combination pattern of the number of local aggregation areas and the diffuse distribution ratio, combined with the coupling coefficient of the electrostatic intensity increase and the hairiness density, the spatial distribution state of the hairiness is analyzed to quantify the degree of dominance of the hairiness density.

[0102] Optionally, when the two-dimensional rating module 302 identifies the type of yarn defect based on the yarn electrostatic state data set and the yarn morphology data collected in real time, and determines the severity of the dominant defect corresponding to the yarn defect type, it is specifically used to: obtain the yarn axial image sequence in real time, and perform the following three-dimensional fluctuation analysis steps according to the yarn axial image sequence: mark the diameter abnormal area where the diameter difference between adjacent sampling points exceeds the diameter tolerance threshold as a sudden uneven area, and synchronously detect the angular change rate of the electrostatic adsorption trajectory of the abnormal diameter area, and mark the current abnormal diameter area when the angular change rate exceeds the preset turbulence threshold. Marked as electrostatic turbulence type sudden unevenness; the trend abnormal area corresponding to the unidirectional continuous increase / decrease diameter trend that exceeds the preset number of points continuously is marked as a gradual unevenness area, and the electrostatic gradient distribution slope of the gradual area is obtained simultaneously. When the slope is in the same direction as the diameter change trend and exceeds the associated threshold, it is marked as electrostatic accumulation type gradual unevenness; determine the proportion of the length of the sudden unevenness area and the length of the gradual unevenness area in the unit detection area, and quantify the degree of dominance of the strip unevenness by combining the maximum diameter deviation value and the electrostatic disorder intensity parameter that are positively correlated with the degree of dominance of the strip unevenness and the electrostatic gradient consistency parameter that is negatively correlated with the degree of dominance of the strip unevenness.

[0103] Optionally, when the two-dimensional rating module 302 identifies the yarn defect type based on the yarn electrostatic state data set and the yarn morphology data collected in real time, and determines the severity of the dominant defect corresponding to the yarn defect type, it is specifically used to: obtain yarn surface scanning imaging data in real time, and synchronously capture the corresponding electrostatic intensity extreme value, and perform the following impurity multi-scale feature extraction steps to clarify the impurity attributes: when there is a dark clumping structure with irregular boundaries whose charge decay rate is less than a preset decay rate threshold, the dark clumping structure is determined to be a fiber agglomerate; when there are reflective particles with geometric edges whose electrostatic intensity extreme value is greater than a preset multiple of the particle size, the reflective particles are determined to be hard impurities; when there is a radial fiber entanglement structure whose electrostatic intensity is in an oscillating and fluctuating state, the radial fiber entanglement structure is determined to be a yarn defect nodule; according to the size grade, optical contrast and boundary clarity of the impurity multi-scale features, the size grade, the optical contrast and the boundary clarity are weighted according to the impurity attributes to quantify the degree of dominant fiber knots.

[0104] Optionally, when the two-dimensional rating module 302 generates a real-time quality rating of the yarn based on the dynamic rating score, and the real-time quality rating of the yarn is divided into three grades: excellent, medium and poor, it is specifically used to: based on a preset rule table, map the identified several yarn defect types to defect type identifiers, and map the assessed severity of the dominant defect to a severity level identifier; perform string splicing on several defect type identifiers and corresponding severity level identifiers, and use the spliced ​​result as a unit rating code; after all the unit rating codes are spliced, perform secondary string splicing on several unit rating codes to generate a composite rating code; according to the preset mapping rules, map the composite rating code to the initial quality of excellent / medium / poor, etc. level; the mapping rules include: determining the total number of all defects as the defect density, and identifying the highest severity level identifier in the composite rating code; if the defect density is greater than a first density threshold or the highest severity level identifier reaches the first severity level threshold, it is judged as poor; if the defect density does not meet the aforementioned condition for being judged as poor but is greater than a second density threshold or the highest severity level identifier reaches the second severity level threshold, it is judged as medium; if the defect density does not exceed the second density threshold and the highest severity level identifier does not reach the second severity level threshold, it is judged as excellent; wherein, the first density threshold is greater than the second density threshold, and the first severity level threshold is higher than the second severity level threshold.

[0105] Optionally, the trend deduction module 303 is specifically used to: extract the trajectory time series dynamic characteristics including trajectory fluctuation frequency, directional consistency and intensity variation amplitude according to the electrostatic adsorption abnormal trajectory; identify the trajectory evolution mode based on the dynamic characteristics: when the trajectory fluctuation frequency is lower than the preset low-frequency threshold and the directional consistency is higher than the stability threshold, it is a steady slowdown mode; when the trajectory fluctuation frequency is higher than the preset high-frequency threshold and the intensity variation amplitude continues to increase, it is a deterioration diffusion mode; when the directional consistency is lower than the disorder threshold and the intensity variation amplitude periodically decreases, it is an intermittent oscillation mode; map the trajectory evolution mode to a quality trend: the steady slowdown mode is mapped to a defect relief trend, the deterioration diffusion mode is mapped to a defect escalation trend, and the intermittent oscillation mode is mapped to a quality fluctuation trend; generate the subsequent quality change trend of the yarn according to the quality trend combined with the severity of the dominant defect.

[0106] Optionally, the control execution module 304 is specifically used to: perform differentiated control according to the subsequent quality change trend type of the yarn: if it is determined to be the defect escalation trend, trigger linkage control, reduce the spinning speed and increase the environmental humidification; if it is determined to be the defect relief trend, maintain the current process and reduce the detection density; if it is determined to be the quality fluctuation trend, dynamically adjust the environmental temperature and humidity and start static electricity neutralization in a targeted manner; associate the real-time quality rating of the yarn, the subsequent quality change trend of the yarn and the control instructions with the production line batch and time, and generate and output a quality rating log.

Claims

1. A real-time quality rating method for yarn textiles, characterized in that: include: Obtain spatiotemporal charge data sets and generate yarn electrostatic state data sets through multimodal sensing fusion strategies; Based on the yarn electrostatic state dataset and the real-time collected yarn morphology data, yarn textile quality defects are identified, and a two-dimensional dynamic rating is performed according to the defect type and severity to generate a real-time yarn quality rating; performing a change trend analysis on the yarn real-time quality rating according to the yarn real-time quality rating to generate a subsequent quality change trend of the yarn; According to the subsequent quality change trend of the yarn, corresponding process control instructions are executed and a quality rating log is output; The multimodal sensing fusion strategy includes: Using a spatially deployed annular array of electrostatic sensors, the spatiotemporal charge dataset at key points of the yarn is collected in real time and contactlessly. The spatiotemporal charge data set includes real-time charge intensity values ​​and real-time charge distribution characteristics; Using high frame rate visual sensors to capture yarn shape data and motion trajectory in real time; Synchronously analyzing the temporal and spatial correlations between the real-time charge intensity value, the real-time charge distribution characteristics, and the yarn morphology data to identify abnormal electrostatic adsorption trajectories caused by quality defects; The yarn electrostatic state dataset is generated according to the spatiotemporal charge dataset, the yarn morphology data and the electrostatic adsorption abnormal trajectory.

2. The method according to claim 1, characterized in that The method of identifying yarn textile quality defects based on the yarn electrostatic state dataset and combining it with the yarn morphology data collected in real time, and performing a two-dimensional dynamic rating according to the defect type and severity to generate an initial yarn quality rating includes: Identifying the type of yarn defect based on the yarn electrostatic state data set and the yarn morphology data collected in real time, and determining the severity of the dominant defect corresponding to the yarn defect type; The yarn defect types include excessive hairiness, uneven yarn evenness and fiber knots; performing normalization processing on the plurality of yarn defect types to a unified numerical range, and determining the normalized yarn defect type; determining a dynamic rating score according to the normalized yarn defect type and the severity of the dominant defect; According to the dynamic rating score, a real-time quality rating of the yarn is generated, and the real-time quality rating of the yarn is divided into three grades: excellent, medium and poor.

3. The method according to claim 2, characterized in that The identifying of the yarn defect type based on the yarn electrostatic state data set and the yarn morphology data collected in real time, and determining the severity of the dominant defect corresponding to the yarn defect type, includes: Acquire the yarn surface image and the corresponding real-time charge intensity value in real time, and perform the following hairiness density gridding analysis steps based on the yarn surface image: When the number of discrete fibers in the detection data for several consecutive times exceeds the preset first density threshold and the average value of their electrostatic intensity is greater than the preset multiple of the visual hairiness number, it is determined to be a local hairiness aggregation defect; When the number of discrete fibers in the unit detection area exceeding the preset ratio exceeds the preset second density threshold and the overall electrostatic distribution unevenness is greater than the preset diffuse hairiness ratio, it is determined to be a diffuse hairiness excess defect; According to the combination pattern of the number of local aggregation areas and the diffuse distribution ratio, combined with the coupling coefficient between the electrostatic intensity increase and the hairiness density, the spatial distribution state of the hairiness is analyzed and the degree of hairiness density dominance is quantified.

4. The method according to claim 3, characterized in that The identifying of the yarn defect type based on the yarn electrostatic state data set and the yarn morphology data collected in real time, and determining the severity of the dominant defect corresponding to the yarn defect type, includes: A yarn axial image sequence is acquired in real time, and the following three-dimensional evenness fluctuation analysis steps are performed based on the yarn axial image sequence: Marking a diameter abnormal area where the diameter difference between adjacent sampling points exceeds a diameter tolerance threshold as a sudden uneven area, and simultaneously detecting the angular change rate of the electrostatic adsorption trajectory within the diameter abnormal area. When the angular change rate exceeds a preset turbulence threshold, marking the current diameter abnormal area as an electrostatic turbulence sudden uneven area; The abnormal trend area corresponding to the unidirectional continuous increase / decrease diameter trend exceeding the preset number of points is marked as a gradual uneven area, and the electrostatic gradient distribution slope of the gradual area is obtained simultaneously. When the slope is in the same direction as the diameter change trend and exceeds the associated threshold, it is marked as electrostatic accumulation type gradual unevenness; The proportion of the length of the sudden uneven area and the length of the gradual uneven area in the unit detection area is determined, and the degree of the uneven stripping dominance is quantified by combining the maximum diameter deviation value and the electrostatic disorder intensity parameter which are positively correlated with the degree of uneven stripping dominance, and the electrostatic gradient consistency parameter which is negatively correlated with the degree of uneven stripping dominance.

5. The method according to claim 4, characterized in that The identifying of the yarn defect type based on the yarn electrostatic state data set and the yarn morphology data collected in real time, and determining the severity of the dominant defect corresponding to the yarn defect type, includes: Real-time acquisition of yarn surface scanning imaging data and simultaneous capture of corresponding electrostatic intensity extremes are performed to perform the following impurity multi-scale feature extraction steps to clarify impurity properties: When there is a dark mass structure with an irregular boundary whose charge decay rate is less than a preset decay rate threshold, the dark mass structure is determined to be a fiber agglomerate; When there are reflective particles with geometric edges and corners whose electrostatic intensity extreme value is greater than a preset multiple of the particle size, the reflective particles are determined to be hard impurities; When there is a radial fiber entanglement structure with an oscillating electrostatic intensity, the radial fiber entanglement structure is determined to be a yarn defect nodule; According to the size grade, optical contrast and boundary clarity of the multi-scale characteristics of the impurities, the size grade, the optical contrast and the boundary clarity are weighted according to the impurity attributes to quantify the degree of fiber knot visibility.

6. The method according to claim 5, characterized in that Generating a real-time yarn quality rating based on the dynamic rating score, wherein the real-time yarn quality rating is divided into three grades: excellent, medium, and poor, including: Based on a preset rule table, the identified yarn defect types are mapped into defect type identifiers, and the assessed severity of the dominant defect is mapped into a severity level identifier; Concatenate the defect type identifiers with the corresponding severity level identifiers, and use the concatenated result as the unit rating code; After all the unit rating codes are concatenated, a secondary string concatenation is performed on a number of the unit rating codes to generate a composite rating code; According to a preset mapping rule, the composite rating code is mapped to an initial quality grade of excellent / medium / poor; The mapping rules include: determining the total number of all defects as the defect density, and identifying the highest severity level identifier in the composite rating code; If the defect density is greater than the first density threshold or the highest severity level indicator reaches the first severity level threshold, it is determined to be poor; If the defect density does not meet the aforementioned condition for being judged as poor but is greater than the second density threshold or the highest severity level indicator reaches the second severity level threshold, it is judged as medium; If the defect density does not exceed the second density threshold and the highest severity level indicator does not reach the second severity level threshold, it is determined to be excellent; The first density threshold is greater than the second density threshold, and the first severity level threshold is higher than the second severity level threshold.

7. The method according to claim 6, characterized in that The method of performing a change trend analysis on the real-time quality rating of the yarn according to the real-time quality rating of the yarn to generate a subsequent quality change trend of the yarn includes: Extracting trajectory temporal dynamic features including trajectory fluctuation frequency, direction consistency, and intensity variation amplitude based on the electrostatic adsorption abnormal trajectory; Identify trajectory evolution patterns based on the dynamic features: When the trajectory fluctuation frequency is lower than the preset low-frequency threshold and the direction consistency is higher than the stability threshold, it is in a steady slowing mode; When the trajectory fluctuation frequency is higher than the preset high-frequency threshold and the intensity variation amplitude continues to increase, it is a deterioration diffusion mode; When the directional consistency is lower than the disorder threshold and the amplitude of intensity variation decreases periodically, it is an intermittent oscillation mode; Mapping the trajectory evolution pattern into a quality trend: the steady slowdown pattern is mapped into a defect alleviation trend, the deterioration diffusion pattern is mapped into a defect escalation trend, and the intermittent oscillation pattern is mapped into a quality fluctuation trend; A subsequent quality change trend of the yarn is generated according to the quality trend and the severity of the dominant defect.

8. The method according to claim 7, characterized in that The method of executing corresponding process control instructions according to the subsequent quality change trend of the yarn and outputting a quality rating log includes: Differentiated control is performed according to the subsequent quality change trend of the yarn: If it is determined that the defect is escalating, the linkage control is triggered to reduce the spinning speed and increase the environmental humidification; If it is determined that the defect is mitigating, the current process is maintained and the inspection density is reduced; If it is determined that the quality fluctuation trend is as described above, the ambient temperature and humidity are dynamically adjusted and static neutralization is initiated in a targeted manner; The real-time quality rating of the yarn, the subsequent quality change trend of the yarn and the control instructions are associated with the production line batch and time, and a quality rating log is generated and output.

9. A real-time quality rating system for yarn textiles, characterized in that: The method as claimed in any one of claims 1 to 8 comprises: The charge monitoring module is used to obtain the spatiotemporal charge data set and generate the yarn electrostatic state data set through a multimodal sensor fusion strategy; A two-dimensional rating module is used to identify yarn textile quality defects based on the yarn electrostatic state data set and the real-time collected yarn morphology data, and perform two-dimensional dynamic rating according to the defect type and severity to generate a real-time yarn quality rating; A trend deduction module is used to analyze the change trend of the yarn real-time quality rating according to the yarn real-time quality rating, and generate the subsequent quality change trend of the yarn; The control execution module is used to execute corresponding process control instructions according to the subsequent quality change trend of the yarn and output a quality rating log.

Images