Textile processing adaptive optimization method based on internet of things

Abstract

This invention relates to the field of textile industrial process control technology, and particularly to an adaptive optimization method for textile processing based on the Internet of Things (IoT). This method collects real-time data on the unwinding levels of each spindle, filters out target spindles with increasing risk trends, and acquires their yarn air pocket images. It extracts the number of air pocket feature points and their drift speed to determine whether the multi-spindle air pocket is in a precursory or stable state. Based on the unwinding state, it formulates strategies such as yarn guide angle compensation, speed reduction, or tension adjustment. Then, it determines the group drift identifier by using the same drift speed index of precursory spindles within the same region, distinguishing between single-spindle anomalies and regional co-drift, and performs differentiated adjustments. Finally, it determines whether to issue an early warning based on the time required to recover to a stable state. This invention moves control forward to the precursory state, achieving single-spindle autonomy and regional group control collaboration. Through closed-loop verification, it significantly reduces yarn breakage rate, hairiness, and energy consumption, improving the stability and intelligence level of textile processing.

Description

Technical Field

[0001] This invention relates to the field of textile industry process control technology, and in particular to an adaptive optimization method for textile processing based on the Internet of Things. Background Technology

[0002] During textile processing, air pockets are formed when yarn is unwound at high speed from a stationary bobbin. The shape of these air pockets directly affects yarn tension stability, breakage rate, and product quality. When yarn is unwound from different levels (top, middle, and bottom) of the bobbin, the air pocket can easily transform from a single air pocket to a double or multi-spindle air pocket. This transformation process is accompanied by early warning signs such as accelerated node drift and increased tension fluctuations. Existing technologies mostly focus on air pocket shape detection or tension feedback control, lacking early identification of multi-spindle air pocket precursors and adaptive linkage optimization based on multi-spindle group behavior. This makes it difficult to distinguish between local anomalies in a single spindle and regional co-drift problems, resulting in coarse control strategies and delayed responses.

[0003] CN116555953A discloses a device for detecting the shape of an air ring in a straight twisting machine and a control method for a straight twisting machine for cord fabric. The device includes an air ring shape sensor; the sensor comprises an infrared emitter and an infrared receiver facing each other. The line connecting the infrared emitter and receiver is perpendicular to the axis of the spindle used for twisting in the straight twisting machine. The line connecting the infrared emitter and receiver is aligned with the point of maximum fluctuation in the shape of the air ring formed during the straight twisting process. During the operation of the straight twisting machine, the yarn forming the air ring triggers the air ring shape sensor twice per revolution. This invention uses an infrared sensor as the air ring shape sensor, and calculates the air ring radius by utilizing the interval between the two triggers of the infrared sensor by the yarn, thereby analyzing the state of the yarn. This air ring shape detection method only requires adding one more set of infrared sensors to achieve non-contact detection of the air ring, resulting in low cost and wide applicability. Therefore, the existing methods have the following problems: Adjusting the yarn feeding speed via PID control based on the measured air circle radius being too large or too small is a typical deviation feedback control, and it only performs single feedback control adjustment for a single spindle position. However, it fails to identify the precursor state of multiple air circles before the formation of double air circles but when the node drift speed is accelerating, and fails to calculate the group drift indicator based on the drift speed of the characteristic points of all target spindle positions with multi-air circle precursor states in the same area. It also fails to distinguish between single spindle anomalies and regional co-drift through the same direction index, and adopt different adjustment strategies for different anomaly types. Summary of the Invention

[0004] To address this, the present invention provides an adaptive optimization method for textile processing based on the Internet of Things, which overcomes the problems in the prior art, such as the lack of early identification of the precursor states of multiple solar terms, the lack of adaptive linkage optimization based on the collective behavior of multiple spindles, and the difficulty in distinguishing between local anomalies of a single spindle and regional co-drift problems, resulting in coarse control strategies and delayed responses.

[0005] To achieve the above objectives, the present invention provides an adaptive optimization method for textile processing based on the Internet of Things, comprising: Step S1: Real-time acquisition of the unwinding level of each spindle position to determine the risk trend of the multi-stage air-stopping ring of the corresponding spindle position; Step S2: The spindles with a risk trend of increasing risk in the multi-stage air circle are identified as target spindles, and yarn air circle images of each target spindle are obtained to determine the number of air circle feature points and the feature point drift speed. Step S3: Determine the unwinding state of the corresponding target spindle position based on the number of gas circle feature points and the drift velocity of the feature points. The unwinding state includes the multi-section gas circle precursor state and the unwinding gas circle stable state. Step S4: Formulate an unwinding strategy for the target spindle position based on the unwinding state. The unwinding strategy includes adjusting the yarn guide angle, adjusting the running speed, and adjusting the tension parameters. Step S5: Determine the group drift identifier based on the drift velocity of the characteristic points of all multi-throttle ring precursor target spindle positions in the same area. The group drift identifier is the same direction index of the drift velocity of the characteristic points of each spindle position. Step S6: Determine the spindle position anomaly type based on the group drift identifier. The spindle position anomaly type includes single spindle anomaly and regional co-drift. Step S7: Adjust the unwinding strategy according to the spindle position anomaly type and execute the adjusted unwinding strategy; Step S8: Determine whether to issue a warning signal based on the recovery time of each target spindle position in restoring the stable state of the unwinding gas ring after implementing the adjusted unwinding strategy.

[0006] As a preferred technical solution for the IoT-based adaptive optimization method for textile processing, step S1, which determines the multi-stage risk trend of a single spindle, includes: The current unwinding level of a single spindle position is compared with the corresponding historical stable unwinding level range, and in response to the current unwinding level entering the unstable unwinding level range, the risk trend of the multi-stage cycle of the spindle position is determined to be an increasing risk trend.

[0007] As a preferred technical solution for the IoT-based adaptive optimization method for textile processing, step S2, the process of determining the number of air sphere feature points, includes: Edge extraction, contour fitting, and curvature analysis are performed on the yarn air pocket image; Local curvature abrupt changes, local contraction points, or segmental inflection points in the gas sphere contour are identified as gas sphere contour feature points. The number of feature points of the gas circle contour is counted and recorded as the number of gas circle feature points.

[0008] As a preferred technical solution for the adaptive optimization method of textile processing based on the Internet of Things, the process of determining the drift velocity of feature points in step S2 includes: Match the feature points of the atmosphere contour at consecutive sampling times, and calculate the positional displacement of the same feature point of the atmosphere contour between consecutive sampling times; The ratio of the positional displacement to the sampling time interval is determined as the rate of positional change of the feature point of the cyclone contour; The maximum rate of change of position of each feature point of the gas sphere contour is recorded as the feature point drift velocity.

[0009] As a preferred technical solution for the adaptive optimization method of textile processing based on the Internet of Things, in step S3, the unwinding state of the corresponding target spindle position is determined based on the number of feature points of the air ring and the drift velocity of the feature points, including: In response to the number of gas circle feature points being greater than 1 and / or the drift velocity of the feature points continuously increasing for a duration exceeding a preset duration, the unwinding state of the corresponding target spindle position is determined to be a multi-section gas circle precursor state. Conversely, the unwinding state of the corresponding target spindle position is determined to be the unwinding gas ring stable state.

[0010] As a preferred technical solution for the IoT-based adaptive optimization method for textile processing, in step S4, a rewinding strategy for the corresponding target spindle position is formulated based on the rewinding state, wherein: In response to the stable state of the unwinding air ring, the unwinding strategy for the target spindle position is to compensate the guide angle to correct the guide trajectory. In response to the precursor state of multiple throttle rings, the unwinding strategy for the target spindle position is to reduce the operating speed or adjust the tension parameters in the direction of suppressing the formation of multiple throttle rings.

[0011] As a preferred technical solution for the IoT-based adaptive optimization method for textile processing, the process of determining the group drift identifier to determine the spindle position anomaly type in steps S5 and S6 includes: Based on the determination that the number of target spindles is no more than 2, the spindle anomaly type is determined to be a single spindle anomaly. Based on the determination that the number of target spindle positions is greater than 2, the proportion of target spindle positions in the same region that are in the precursor state of multiple throttle cycles with the same direction of drift velocity change is statistically analyzed. The spindle position anomaly type is then determined according to the relationship between this proportion and a preset proportion, wherein: In response to the same-direction ratio being greater than a preset ratio, the same-direction index is determined to be high and the spindle position anomaly type is determined to be regional co-drift. In response to the fact that the same direction ratio is not greater than a preset ratio, the same direction index is determined to be low and the spindle position anomaly type is determined to be a single spindle anomaly.

[0012] As a preferred technical solution for the IoT-based adaptive optimization method for textile processing, in step S7, adjusting the unwinding strategy according to the spindle position anomaly type includes: In response to a single ingot anomaly, the operating speed of the target ingot position is reduced; In response to regional co-drift, the following adjustments will be made to all target spindles within the region that are in a multi-seasonal precursor state, including: The maximum operating speed of each target spindle position is uniformly reduced and the preset time is shortened.

[0013] As a preferred technical solution for the adaptive optimization method of textile processing based on the Internet of Things, in step S8, a warning signal is issued based on the fact that the recovery time of each target spindle position that performs the adjusted unwinding strategy to restore the stable state of the unwinding air ring is greater than the upper limit of the recovery time.

[0014] Compared with existing technologies, the beneficial effects of this invention are that, through multi-level and multi-dimensional collaborative control, it achieves full-process adaptive optimization from single-spindle air circle precursor state identification to regional group anomaly attribution, and then to differentiated strategy adjustment and closed-loop verification; this method moves the control point from the already occurred break-off or obvious multi-air circle to the multi-air circle precursor state, making full use of the edge computing capabilities of the Internet of Things, upgrading the traditional single-spindle open-loop control to a smart optimization method of multi-spindle collaboration, precursor identification, and closed-loop verification, which improves the stability of textile processing and product quality while reducing energy consumption and manual intervention costs, and has significant industrial practical value; In particular, by predicting risks through unwinding layers, high-precision image acquisition is initiated only for spindles in unstable regions (such as the upper and lower parts). By combining the number of characteristic points (nodes) of the air ring with the duration of continuous increase in the drift speed of the characteristic points and determining the precursor state of multiple air rings, early warnings can be issued and interventions can be made before the yarn forms obvious double air rings or breaks. This transforms passive post-event handling into proactive pre-event prevention and control, effectively reducing yarn breaks, hairiness and poor packaging, and improving the quality of finished products. In particular, by utilizing the unidirectional drift velocity index of multiple precursor state spindles within the same region, a group drift signature is constructed: when the number of precursor state spindles is greater than 2 and the unidirectional ratio exceeds a preset threshold, it is determined to be regional co-drift (environmental or batch problem); otherwise, it is a single spindle anomaly. Differentiated strategies are implemented for different anomaly types, i.e., single spindle anomalies only result in local speed reduction to avoid affecting other spindles, while regional co-drift uniformly reduces the maximum operating speed of all precursor state spindles within the region and shortens the preset time for precursor state determination to improve sensitivity. This avoids mistaking group problems for single-point failures and thus ineffective intervention, and also prevents sporadic disturbances from being excessively amplified into global speed reduction, significantly improving the robustness and economy of control. In particular, after implementing the adjusted unwinding strategy, the recovery time required for each target spindle position to return to the stable state of the unwinding gas ring is monitored. If the recovery time exceeds the preset upper limit, the intervention effect is determined to be insufficient and an early warning signal is issued, prompting the operator or triggering a higher level of fault investigation (such as pulse speed reduction, ring suppression intervention, equipment isolation). This mechanism ensures the effectiveness of each control action, prevents continuous deterioration due to control failure, and provides traceable process data for the IoT platform, facilitating remote operation and maintenance and model iteration. Attached Figure Description

[0015] Figure 1 This is a flowchart illustrating the steps of an adaptive optimization method for textile processing based on the Internet of Things, according to an embodiment of the present invention. Detailed Implementation

[0016] To make the objectives and advantages of the present invention clearer, the present invention will be further described below with reference to embodiments; it should be understood that the specific embodiments described herein are merely for explaining the present invention and are not intended to limit the present invention.

[0017] Preferred embodiments of the present invention will now be described with reference to the accompanying drawings. Those skilled in the art should understand that these embodiments are merely illustrative of the technical principles of the present invention and are not intended to limit the scope of protection of the present invention.

[0018] It should be noted that in the description of this invention, the terms "upper", "lower", "left", "right", "inner", "outer", etc., which indicate directions or positional relationships, are based on the directions or positional relationships shown in the accompanying drawings. This is only for the convenience of description and is not intended to indicate or imply that the device or element must have a specific orientation, or be constructed and operated in a specific orientation. Therefore, it should not be construed as a limitation of this invention.

[0019] Furthermore, it should be noted that, in the description of this invention, unless otherwise explicitly specified and limited, the terms "installation," "connection," and "linking" should be interpreted broadly. For example, they can refer to a fixed connection, a detachable connection, or an integral connection; they can refer to a mechanical connection or an electrical connection; they can refer to a direct connection or an indirect connection through an intermediate medium; and they can refer to the internal connection of two components. Those skilled in the art can understand the specific meaning of the above terms in this invention according to the specific circumstances.

[0020] Please see Figure 1 The diagram shown illustrates the steps of an adaptive optimization method for textile processing based on the Internet of Things, according to an embodiment of the present invention.

[0021] This invention provides an adaptive optimization method for textile processing based on the Internet of Things, comprising: Step S1: Collect the unwinding level of each spindle position in real time, compare the current unwinding level of a single spindle position with the corresponding historical stable unwinding level range, and determine the risk trend of the multi-stage ring of the spindle position as an increasing risk trend when the current unwinding level enters the unstable unwinding level range, and otherwise determine it as a stable risk trend. Understandably, when yarn is pulled out of a stationary bobbin at high speed, a rotating air ring is formed: In the upper unwinding stage (i.e., when more yarn remains), the unwinding point is close to the top of the bobbin, and the angle between the yarn and the guide hook when the yarn leaves the bobbin is close to 90°, resulting in a large initial air ring radius and strong centrifugal force. At this time, the air ring is very prone to outward expansion. Once the tension fluctuates slightly, it will change from a single air ring to a double air ring or even a multi-section air ring. Therefore, the upper region is called the high-energy zone, and the air ring shape is naturally unstable. In the lower unwinding stage (i.e., when less yarn remains), the unwinding point is close to the bottom of the bobbin, and the yarn needs to wrap around the bobbin. The unwinding path of the yarn can only enter the guide hook from the shoulder, which is winding and the friction between the yarn and the surface of the package yarn is intensified, resulting in periodic fluctuations in tension. At the same time, the rapid change of the unwinding point will cause the bottom boundary conditions of the loop to change continuously, which can easily induce standing wave nodes, thus forming double loops or multi-node loops. Therefore, the lower area is called the friction zone and is also a high-instability zone. The unwinding point of the middle unwinding is located in the middle of the package yarn, the yarn departure angle is moderate, the unwinding path is smooth, and the geometry of the loop naturally forms a stable single loop. Moreover, the unwinding point moves slowly and is not easy to induce node splitting. Therefore, the middle is a stable zone. Typically, all unwinding level values ​​that show stable single air loops are statistically analyzed, and the lower and upper limits of the 95% confidence interval are taken as the historical stable unwinding level interval. In one implementation, the historical stable unwinding level interval is when the remaining yarn height accounts for 30% to 70% of the initial package height, while the unstable unwinding level interval includes the upper unstable area and the lower unstable area. The upper unstable area, i.e., the upper part of the yarn package, usually has a remaining height greater than 70%, and the lower unstable area, i.e., the bottom of the yarn package, usually has a remaining height less than 30%. It should be understood that textile engineering has long confirmed that the breakage rate and unwinding level follow a bathtub curve (i.e., high at both ends and low in the middle), which indicates that the unwinding level is an important characteristic of the gas circle's stability. Currently, the unwinding level is in the range that has been statistically proven to be most prone to producing multi-segment gas circles in history. Even if the gas circle has not yet shown multi-segment gas circles, since the unwinding level has entered the unstable zone, it can be predicted a priori that the gas circle morphology will deteriorate in a short period of time if no intervention measures are taken.

[0022] Step S2: The spindles with a risk trend of increasing risk in the multi-stage air circle are identified as target spindles, and yarn air circle images of each target spindle are obtained to determine the number of air circle feature points and the feature point drift speed. Specifically, edge extraction, contour fitting, and curvature analysis are performed on the yarn airfoil image; local curvature abrupt change points, local contraction points, or segmented turning points in the airfoil contour are identified as airfoil contour feature points; the number of airfoil contour feature points is counted and recorded as the airfoil feature point count; airfoil contour feature points at consecutive sampling times are matched, and the positional displacement of the same airfoil contour feature point between consecutive sampling times is calculated; the ratio of the positional displacement to the sampling time interval is determined as the positional change rate of the airfoil contour feature point; the maximum positional change rate of each airfoil contour feature point is recorded as the feature point drift velocity. In implementation, an air bubble image acquisition system is installed at each spindle position of textile equipment (such as winding machines and twisting machines). The core hardware includes: a CCD industrial camera, an auxiliary light source (LED ring light source or backlight), and an industrial lens selected according to the actual size of the yarn air bubble (generally with a diameter of tens to hundreds of millimeters) and the distance between the camera and the yarn. The industrial camera is mounted on the side of the yarn air bubble with its optical axis perpendicular to the air bubble axis (i.e., horizontally aligned with the position of maximum expansion of the air bubble) to acquire a side-view projection image of the air bubble on a two-dimensional plane. It can be understood that image acquisition is triggered only at the target spindle position (i.e., the spindle position that has been determined to have an increasing risk trend). In one implementation, after acquiring the original yarn balloon image, preprocessing operations are first performed (including denoising, image enhancement, grayscale conversion, and binarization). Edge detection is then performed on the preprocessed image to extract the contour of the yarn balloon (including Canny, Sobel, and LoG operators). After extracting the edge point set, the discrete edge points are fitted into a continuous smooth curve (B-spline curves or polynomial fitting can be used). Subsequently, curvature analysis is performed on the fitted contour curve to determine the curvature value of each point. Local curvature abrupt change points, local contraction points, and segmented inflection points are identified and determined as balloon contour feature points. All identified feature points are counted to obtain the number of balloon feature points. It should be understood that the physical essence of feature point drift velocity is the movement rate of standing wave nodes on the yarn air ring in space, reflecting the drastic degree of air ring morphological change. In air ring dynamics, the node positions of stable single air rings have relatively low (even close to zero) drift velocities compared to fixed nodes. When the air ring is in the critical region of transition from a single air ring to a double air ring, the stability of the nodes begins to decrease, and they drift along the air ring contour, with the drift velocity gradually increasing. Air ring instability usually starts from the most unstable node, and the drift velocity of this node best reflects the degree of air ring morphological deterioration. Taking the maximum value can maximize the sensitivity of precursor state recognition and avoid obscuring key information due to averaging. In implementation, feature point matching and drift velocity calculation can be performed using the sparse optical flow method. The optical flow method calculates the instantaneous velocity of pixel movement, and by analyzing the changes in pixel points, it represents the unique motion of the image, thereby achieving target tracking. For a small number of feature points in a yarn air ring (usually no more than 5), the sparse optical flow method has high computational efficiency and can meet real-time requirements.

[0023] Step S3, determining the unwinding state of the corresponding target spindle position based on the number of characteristic points of the gas ring and the drift velocity of the characteristic points, includes: In response to the number of gas circle feature points being greater than 1 and / or the drift velocity of the feature points continuously increasing for a duration exceeding a preset time, the unwinding state of the corresponding target spindle position is determined to be a multi-segment gas circle precursor state. It can be understood that the number of feature points ≥ 2 represents that the gas circle has the geometric structure of a multi-segment gas circle (spatial instability), the continuous increase in drift velocity represents that the gas circle is in a dynamic deterioration trend (temporal instability), and if the velocity continuously increases for a duration exceeding a preset time, it indicates that the gas circle is in a dynamic deterioration trend of transitioning from stability to instability. Therefore, if any condition is met, it is determined to be a multi-segment gas circle precursor state, realizing a dual-channel early warning of spatial structure and temporal trend. Conversely, the unwinding state of the corresponding target spindle position is determined to be the unwinding gas ring stable state; In practice, the drift speed of the yarn nodes during high-speed movement may exhibit random spikes due to airflow disturbances, uneven yarn twist, and other reasons. If the speed increase at a single sampling point is judged as a precursor state, it will lead to frequent false triggering. The time for the yarn to develop from a stable single air loop to a clear double air loop is usually between 0.5 seconds and 2 seconds. The precursor state should appear in the early stage of this process. Therefore, the preset duration is set to 0.3 seconds to 0.5 seconds. Preferably, the preset duration is set to 0.3 seconds. At this time, there is still an intervention window of about 0.2 seconds before the actual formation of the double air loop, which is sufficient to perform actions such as deceleration or tension adjustment.

[0024] Step S4: Based on the unwinding state, formulate an unwinding strategy for the corresponding target spindle position, wherein: In response to the stable state of the unwinding air ring, the unwinding strategy for the target spindle position is to compensate the guide angle to correct the guide trajectory. It can be understood that the stable state of the unwinding air ring indicates that although the unwinding level is in the unstable zone, the yarn is in or close to a stable single air ring shape, and there may only be a small geometric deviation (such as guide trajectory deviation). At this time, the risk is low, and there is no need to change the operating parameters significantly. By compensating the guide angle and finely adjusting the initial direction of the yarn entering the air ring, the trajectory deviation can be corrected to maintain the stable state, and the control cost is minimal and does not affect production efficiency. In response to the precursory state of multiple air loops, the unwinding strategy for the target spindle position is to reduce the running speed or adjust the tension parameters in the direction of suppressing the formation of multiple air loops. It can be understood that the precursory state of multiple air loops indicates that the air loop is transitioning from a single air loop to a double air loop / multiple air loop. At this time, the guide angle compensation is not enough to suppress instability, and stronger intervention is required: including reducing the running speed to directly reduce the centrifugal force, causing the air loop diameter to shrink and reducing the driving force of node splitting; and adjusting the tension parameters in the direction of suppressing multiple air loops (such as appropriately increasing the tension) to increase the yarn stiffness and suppress the formation of standing wave nodes. Both of these measures block the deterioration process from the perspective of energy and boundary conditions, preventing the development of a complete multiple air loop that leads to yarn breakage.

[0025] Step S5: Determine the group drift identifier based on the drift velocity of the characteristic points of all multi-throttle ring precursor target spindle positions in the same area. The group drift identifier is the same direction index of the drift velocity of the characteristic points of each spindle position. It is understandable that the drift velocity change of a single spindle may be caused by local factors (such as yarn tube deformation or sensor noise at that spindle), while the simultaneous occurrence of velocity changes in the same direction at multiple spindles indicates the existence of common external driving factors. Therefore, by statistically analyzing whether the drift velocity change directions of multiple precursor spindles in the same region are consistent, a unidirectional index is constructed as the core criterion for distinguishing between local problems and group problems. Step S6, determining the spindle position anomaly type based on the group drift identifier, including: Based on the determination that the number of target spindles is no more than 2, the spindle anomaly type is determined to be single spindle anomaly. It is understandable that when the number of spindles in the region in the multi-seasonal precursor state is ≤2, the statistical significance is insufficient, and it is directly determined to be single spindle anomaly (local problem) because the sample is too small to reliably determine whether it is a group co-drift, and even if there is a co-drift trend, the impact is limited and it is simpler to treat it as a single spindle. Based on the determination that the number of target spindles is greater than 2, the proportion of the drift velocity change direction of all target spindles in the same region that are in the multi-throttle ring precursor state is statistically analyzed, and the spindle anomaly type is determined according to the relationship between the proportion of the same direction and the preset proportion. Understandably, when the number of precursor state spindles is greater than 2, the proportion of spindles with the same drift velocity change direction is calculated (same direction ratio). If the same direction ratio is greater than the preset ratio, it means that most spindles deteriorate synchronously and are judged as regional co-drift (batch or environmental problem); otherwise, it is still judged as multiple independent single spindle anomalies. In practice, the preset ratio is usually 50% to 70%. When the ambient temperature and humidity drift slowly or the overall twist of the batch is too large, more than 80% of the spindles in the same area will usually show similar behavior. Setting it to 60% can cover most of the co-drift scenarios while excluding less than half of the accidental coincidences.

[0026] Step S7, adjusting and executing the unwinding strategy according to the spindle position anomaly type, including: In response to a single spindle anomaly, the operating speed of that target spindle is reduced (typically by 5%–15%). This is understandable because a single spindle anomaly indicates a problem originating from localized factors within that spindle (such as yarn bobbin eccentricity, sensor drift, or localized yarn accumulation), rather than an environmental or batch-wide issue. Reducing the spindle's operating speed in this case decreases centrifugal force, causing the air ring diameter to shrink, suppressing node splitting and accelerated drift, without affecting the production efficiency of other normal spindles. Furthermore, a single spindle anomaly does not alter the overall process window for the region, eliminating the need to adjust global parameters (such as maximum operating speed or judgment sensitivity). In response to regional co-drift, the following adjustments will be made to all target spindles within the region that are in a multi-seasonal precursor state, including: The maximum operating speed of each target spindle position is uniformly reduced (operating speed is usually reduced by 15% to 25%) and the preset time is shortened (shortened by 20% to 50%). It is understandable that regional co-drift indicates that multiple spindle positions are showing a deterioration trend simultaneously. The root cause is usually environmental disturbance (temperature and humidity changes, wind field fluctuations) or batch problems (high raw material twist, drift of previous processes). Understandably, when a single ingot is abnormal, the problem originates locally, and a slight reduction in speed can disrupt the conditions for node formation, thus avoiding excessive impact on efficiency; when there is regional co-drift, there is a continuous external driving force, requiring a more significant reduction in speed to quickly reduce the gas ring energy below the stable threshold. In implementation, uniformly reducing the maximum operating speed weakens centrifugal force at the source and suppresses the expansion of the gas ring in all affected spindles. This is a global and preventative group control measure. In addition, the preset duration is the time window for determining the continuous increase in the drift speed of the characteristic point. Shortening this duration means becoming more sensitive to the continuous increase in drift speed. That is, environmental or batch problems may persist in the context of regional co-drift, and it is necessary to identify the spindles that subsequently deteriorate more quickly so as to intervene in a timely manner. Shortening the preset duration is equivalent to improving the monitoring sensitivity.

[0027] Step S8: If the recovery time of each target spindle position after implementing the adjusted unwinding strategy to restore the stable state of the unwinding gas ring is greater than the upper limit of the recovery time, a warning signal is issued; otherwise, no warning signal is issued. Understandably, if no new precursor states appear within the area for a continuous period of time (≤ the upper limit of recovery time) and all existing spindles have recovered to a stable state, the maximum operating speed and preset time will be automatically restored to normal values ​​to avoid long-term speed limits affecting production capacity. In practice, the upper limit of recovery time is usually 5 seconds (the time for yarn to develop from a stable single air ring to a clear double air ring / multi-section air ring is usually 0.5 seconds to 2 seconds. After the speed reduction or tension adjustment command is issued, the motor response, yarn tension rebalancing, and air ring shape adjustment to stability usually take 1 to 3 seconds. Therefore, the air ring should be observed to recover to stability within 3 to 4 seconds. Considering the equipment response delay, yarn tension balancing time, and a certain safety margin, the upper limit of recovery time is set to 1.5 times the normal adjustment time).

[0028] The technical solution of the present invention has been described above with reference to the preferred embodiments shown in the accompanying drawings. However, it will be readily understood by those skilled in the art that the scope of protection of the present invention is obviously not limited to these specific embodiments. Without departing from the principles of the present invention, those skilled in the art can make equivalent changes or substitutions to the relevant technical features, and the technical solutions after these changes or substitutions will all fall within the scope of protection of the present invention.

[0029] The above description is merely a preferred embodiment of the present invention and is not intended to limit the invention. Various modifications and variations can be made to the present invention by those skilled in the art. Any modifications, equivalent substitutions, improvements, etc., made within the spirit and principles of the present invention should be included within the scope of protection of the present invention.

Claims

1. An adaptive optimization method for textile processing based on the Internet of Things, characterized in that, include: Real-time data collection of unwinding levels at each spindle position to determine the risk trend of multiple air-stopping rings at the corresponding spindle position; The spindle positions with a risk trend of increasing risk in the multi-stage air ring are identified as target spindle positions, and yarn air ring images of each target spindle position are obtained to determine the number of air ring feature points and the feature point drift speed. The unwinding state of the corresponding target spindle position is determined based on the number of gas circle feature points and the drift velocity of the feature points. The unwinding state includes the multi-section gas circle precursor state and the unwinding gas circle stable state. Based on the unwinding state, a corresponding unwinding strategy is formulated for the target spindle position. The unwinding strategy includes adjusting the guide angle, adjusting the running speed, and adjusting the tension parameters. The group drift identifier is determined based on the drift velocity of the characteristic points of all multi-stage gas circle precursor target spindle positions in the same area. The group drift identifier is an index of the unidirectional drift velocity of the characteristic points of each spindle position. The spindle position anomaly type is determined based on the group drift identifier, and the spindle position anomaly type includes single spindle anomaly and regional co-drift; Adjust the unwinding strategy according to the type of spindle position abnormality, and execute the adjusted unwinding strategy; Whether to issue an early warning signal is determined based on the recovery time of each target spindle position in restoring the stable state of the unwinding gas ring after implementing the adjusted unwinding strategy.

2. The adaptive optimization method for textile processing based on the Internet of Things according to claim 1, characterized in that, The process of determining the multi-stage risk trend of a single spindle includes: The current unwinding level of a single spindle position is compared with the corresponding historical stable unwinding level range, and in response to the current unwinding level entering the unstable unwinding level range, the risk trend of the multi-stage cycle of the spindle position is determined to be an increasing risk trend.

3. The adaptive optimization method for textile processing based on the Internet of Things according to claim 1, characterized in that, The process of determining the number of atmospheric feature points includes: Edge extraction, contour fitting, and curvature analysis are performed on the yarn air pocket image; Local curvature abrupt changes, local contraction points, or segmental inflection points in the gas sphere contour are identified as gas sphere contour feature points. The number of feature points of the gas circle contour is counted and recorded as the number of gas circle feature points.

4. The adaptive optimization method for textile processing based on the Internet of Things according to claim 3, characterized in that, The process of determining the drift velocity of feature points includes: Match the feature points of the atmosphere contour at consecutive sampling times, and calculate the positional displacement of the same feature point of the atmosphere contour between consecutive sampling times; The ratio of the positional displacement to the sampling time interval is determined as the rate of positional change of the feature point of the cyclone contour; The maximum rate of change of position of each feature point of the gas sphere contour is recorded as the feature point drift velocity.

5. The adaptive optimization method for textile processing based on the Internet of Things according to claim 1, characterized in that, Determining the unwinding state of the corresponding target spindle position based on the number of gas ring feature points and the drift velocity of the feature points includes: In response to the number of gas circle feature points being greater than 1 and / or the drift velocity of the feature points continuously increasing for a duration exceeding a preset duration, the unwinding state of the corresponding target spindle position is determined to be a multi-section gas circle precursor state. Conversely, the unwinding state of the corresponding target spindle position is determined to be the unwinding gas ring stable state.

6. The adaptive optimization method for textile processing based on the Internet of Things according to claim 5, characterized in that, Based on the unwinding state, a corresponding unwinding strategy for the target spindle position is formulated, wherein: In response to the stable state of the unwinding air ring, the unwinding strategy for the target spindle position is to compensate the guide angle to correct the guide trajectory. In response to the precursor state of multiple throttle rings, the unwinding strategy for the target spindle position is to reduce the operating speed or adjust the tension parameters in the direction of suppressing the formation of multiple throttle rings.

7. The adaptive optimization method for textile processing based on the Internet of Things according to claim 1, characterized in that, The process of identifying group drift markers to determine the type of spindle anomaly includes: Based on the determination that the number of target spindles is no more than 2, the spindle anomaly type is determined to be a single spindle anomaly. Based on the determination that the number of target spindle positions is greater than 2, the proportion of target spindle positions in the same region that are in the precursor state of multiple throttle cycles with the same direction of drift velocity change is statistically analyzed. The spindle position anomaly type is then determined according to the relationship between this proportion and a preset proportion, wherein: In response to the same-direction ratio being greater than a preset ratio, the same-direction index is determined to be high and the spindle position anomaly type is determined to be regional co-drift. In response to the fact that the same direction ratio is not greater than a preset ratio, the same direction index is determined to be low and the spindle position anomaly type is determined to be a single spindle anomaly.

8. The adaptive optimization method for textile processing based on the Internet of Things according to claim 1, characterized in that, Adjusting the unwinding strategy according to the type of spindle position anomaly includes: In response to a single ingot anomaly, the operating speed of the target ingot position is reduced; In response to regional co-drift, the following adjustments will be made to all target spindles within the region that are in a multi-seasonal precursor state, including: The maximum operating speed of each target spindle position is uniformly reduced and the preset time is shortened.

9. The adaptive optimization method for textile processing based on the Internet of Things according to claim 1, characterized in that, If the recovery time of each target spindle position in the adjusted unwinding strategy to the stable state of the unwinding gas ring exceeds the upper limit of the recovery time, an early warning signal will be issued.

Images