A yarn run control method and system

Abstract

The present application relates to the technical field of yarn running control, and particularly relates to a yarn running control method and system. The method comprises the following steps: extracting yarn doubling process failure records from a doubling machine control center, and combining tension fluctuation data obtained by a sensor to sample the records to form a doubling failure sampling sample; then, yarn twist unevenness derivation is performed based on the doubling failure sampling sample to obtain yarn twist unevenness data, and yarn adaptive tension control adjustment is completed according to the data to output tension control adjustment data; finally, the tension control adjustment data is optimized, and data coding and packaging are completed to generate hierarchical control coding data packets, which are sent to the doubling machine control center. The yarn running control technology is more perfect through the optimization of the yarn running control technology.

Description

Technical Field

[0001] This invention relates to the field of yarn running control technology, and in particular to a yarn running control method and system. Background Technology

[0002] Currently, there is a growing demand for coarse denier bright nylon filaments (typically 500D-1200D) with high gloss, a rugged texture, and high abrasion resistance in the fields of bag webbing, outdoor equipment, and high-end apparel accessories. However, when processing them using existing conventional doubling technology, the smooth surface and high rigidity of coarse denier nylon filaments lead to unstable unwinding loops and violent instantaneous tension fluctuations, resulting in uneven twisting and loose twisting in the final product. Summary of the Invention

[0003] Therefore, it is necessary to provide a yarn operation control method and system to solve at least one of the above-mentioned technical problems.

[0004] To achieve the above objective, a yarn running control method is provided, the method comprising the following steps: Step S1: Extract yarn twisting process failure records from the control center of the twisting machine. The yarn twisting process failure records include tension fluctuation data obtained from the tension sensor. Sample the yarn twisting process failure records to obtain a twisting failure sampling sample. Step S2: Based on the failed twisting sample, deduce the yarn twist unevenness to obtain yarn twist unevenness data; based on the yarn twist unevenness data, perform adaptive tension control adjustment of the yarn and output tension control adjustment data; Step S3: Optimize tension control adjustment data, then encode and package the data to output graded control coded data packets, and send them to the control center of the twisting machine to execute yarn running control.

[0005] The present invention also provides a yarn running control system for executing the above-described yarn running control method, the yarn running control system comprising: The sample sampling module is used to extract yarn twisting process failure records from the control center of the twisting machine. The yarn twisting process failure records include tension fluctuation data obtained from sensors. The module performs record sampling on the yarn twisting process failure records to obtain twisting failure sampling samples. The graded control module is used to deduce the yarn twist unevenness based on the failed twisting sample, and obtain the yarn twist unevenness data; based on the yarn twist unevenness data, it performs adaptive tension control adjustment of the yarn and outputs tension control adjustment data; The encoding module is used to optimize tension control adjustment data, and then encode and package the data to output graded control encoding data packets, which are then sent to the control center of the twisting machine to execute yarn running control.

[0006] The beneficial effects of this invention are as follows: It captures records of twisting process failures containing raw sensor data on tension fluctuations from the twisting machine control center, and obtains representative twisting failure samples through scientific sampling. This overcomes the drawbacks of traditional manual fault diagnosis, such as low efficiency, biased data selection, and large subjective errors. It can completely preserve the true tension change characteristics when the yarn process fails, laying a real and reliable data foundation for subsequent fault analysis and parameter derivation. Secondly, based on the sampled samples, it performs logical derivation of yarn twist unevenness, accurately generates twist unevenness data, and then completes yarn adaptive tension control adjustment and outputs exclusive tension control adjustment data. This breaks the limitation of traditional fixed tension parameters being unable to adapt to different yarn materials, thicknesses, and operating conditions. It can specifically solve production pain points such as uneven twist, twisting process failure, yarn breakage, and inconsistent finished product quality caused by abnormal tension, achieving dynamic adaptive matching of tension parameters and significantly improving the uniformity and production stability of yarn twisting. Finally, the tension control adjustment data is deeply optimized, and then a hierarchical control code data package is generated through standardized coding and packaging, which is then sent to the equipment control center for execution and control. This achieves standardized and structured transmission of control data, ensuring efficient, accurate, and unbiased data parsing and execution in the equipment control system. It also forms a standardized fault handling and closed-loop control mechanism, effectively reducing equipment downtime due to process failures, minimizing raw material waste and manual maintenance costs, improving the automation and intelligent control level of the twisting machine, stabilizing the quality of finished yarn, and comprehensively improving the production efficiency, product qualification rate, and industrial intelligent production adaptability of the textile twisting process. Therefore, this invention is an optimization of a traditional yarn operation control method, solving the problems of unstable unwinding loops, violent instantaneous tension fluctuations, and uneven twisting in the final product, thus improving product quality. Attached Figure Description

[0007] Figure 1 A flowchart illustrating the steps of a yarn operation control method; Figure 2 for Figure 1 A detailed flowchart illustrating the implementation steps of step S2. Figure 3 This is a physical diagram of the tension sensor installed in this invention; Figure 4 This is a physical image of the doubling twister used in this invention; Figure 5 This is an example diagram of the interactive interface of the twisting machine in this invention. Detailed Implementation

[0008] The technical method of the present invention will now be clearly and completely described with reference to the accompanying drawings. Obviously, the described embodiments are only some, not all, of the embodiments of the present invention. All other embodiments obtained by those skilled in the art based on the embodiments of the present invention without inventive effort are within the scope of protection of the present invention.

[0009] Furthermore, the accompanying drawings are merely illustrative of the invention and are not necessarily drawn to scale. The same reference numerals in the drawings denote the same or similar parts, and therefore repeated descriptions of them will be omitted. Some block diagrams shown in the drawings are functional entities and do not necessarily correspond to physically or logically independent entities. These functional entities can be implemented in software, in one or more hardware modules or integrated circuits, or in different network and / or processor methods and / or microcontroller methods.

[0010] It should be understood that although the terms "first," "second," etc., may be used herein to describe various units, these units should not be limited by these terms. These terms are used merely to distinguish one unit from another. For example, without departing from the scope of the exemplary embodiments, a first unit may be referred to as a second unit, and similarly, a second unit may be referred to as a first unit. The term "and / or" as used herein includes any and all combinations of one or more of the associated listed items.

[0011] To achieve the above objectives, please refer to Figures 1 to 5 This invention provides a yarn running control method, the method comprising the following steps: Step S1: Extract yarn twisting process failure records from the control center of the twisting machine. The yarn twisting process failure records include tension fluctuation data obtained from the sensors. Sample the yarn twisting process failure records to obtain a twisting failure sampling sample. Step S2: Based on the failed twisting sample, deduce the yarn twist unevenness to obtain yarn twist unevenness data; based on the yarn twist unevenness data, perform adaptive tension control adjustment of the yarn and output tension control adjustment data; Step S3: Optimize tension control adjustment data, then encode and package the data to output graded control coded data packets, and send them to the control center of the twisting machine to execute yarn running control.

[0012] In this embodiment of the invention, reference is made to Figure 1 The above is a schematic flowchart of a yarn running control method according to the present invention. In this example, the yarn running control method includes the following steps: Step S1: Extract yarn twisting process failure records from the control center of the twisting machine. The yarn twisting process failure records include tension fluctuation data obtained from the sensors. Sample the yarn twisting process failure records to obtain a twisting failure sampling sample. In one implementation of this invention, the specific process of retrieving yarn twisting process failure records from the control center of the twisting machine includes calling the equipment operation log, synchronously retrieving tension process data, and screening failure condition markers.

[0013] Specifically, tension sensors are pre-deployed in the contact area between the yarn and the bobbin of the doubling machine. The tension sensors are as follows: Figure 3 As shown, the sensor sampling frequency is set to 2000 times / second, converting the continuous tension analog signal into a digital tension sequence, which is then stored in real time in the process database of the control center. The control center simultaneously records spindle speed, spindle belt running status, wire coil running position, and winding angle changes. For details, please refer to [link to relevant documentation]. Figure 5 As shown in the partial content, yarn tension corresponds to the "tension fluctuation data obtained by the sensor" in the method, which is the core input for subsequent "twist imbalance derivation" and "adaptive tension control adjustment"; yarn stroke and bobbin diameter correspond to "winding angle change information", which are important bases for judging "abnormal surface ripples or forming deviation of the bobbin"; bobbin speed, i.e., "spindle speed", is a key parameter for calculating "spindle speed" and analyzing "tension instability"; diameter increase reflects the change in winding layer thickness, which is related to the mentioned "winding layer thickness correction"; T1 exceedance count corresponds to the counting of failure conditions such as "yarn breakage and sudden drop in twist" in the method, which is a direct manifestation of "automatically generating corresponding failure identification records". When the system detects yarn breakage, sudden drop in twist, abnormal surface ripples of the bobbin, or forming deviation exceeding the preset process tolerance value, it automatically generates corresponding failure identification records. It should be noted that the preset process tolerance value is not fixed, but is obtained by statistically analyzing the operating parameters of qualified products over the past 30 days. The threshold is the average winding offset of qualified batches plus twice the standard deviation, which is the preset process tolerance value. For example, if the average winding offset is 0.42mm and the standard deviation is 0.08mm, then the process tolerance value is set to 0.58mm. Subsequently, a sample of failure records is recorded using a stratified time-period sampling method. The equipment is divided into three levels according to the operating shift: morning shift, afternoon shift, and evening shift. Within each level, 20 consecutive failure segments are extracted according to the frequency of failure. Each failure segment extracts a data window of 15 seconds before and 5 seconds after the failure, thus forming a double twist failure sampling sample.

[0014] Specifically, if a fault record shows that the yarn breakage occurred at 3280 seconds, then the complete tension fluctuation data between 3265 seconds and 3285 seconds is extracted, along with the corresponding spindle speed and belt tension information, to ensure that the subsequent derivation of twist imbalance has a complete time correlation basis. It should be noted that, to eliminate misjudgments caused by occasional mechanical vibrations, the tension fluctuation data needs to be preprocessed with median filtering before sampling, using a 3-point sliding window to remove isolated spikes, and outputting the failed twisting sample.

[0015] Please see Figure 4 The yellow box indicates the area where the yarn is wound and formed during the process, which is also the location where the "abnormal surface ripples and forming deviation" mentioned in the description occur; the blue box indicates the installation of... Figure 3 The tension sensor shown is located in the area where the yarn contacts the bobbin. The spindle and traveler operating status are also monitored synchronously in this area. The red box represents the yarn supply area, providing the raw yarn for the doubling process. During operation, the yarn from the raw yarn supply area (red box) is drawn out and passes through the blue box, the tension detection and process control area. The tension sensor in this area collects tension fluctuation data at a frequency of 2000 times per second and simultaneously monitors key parameters such as spindle speed, spindle status, and traveler position. The yarn is then pulled to the yellow box, the bobbin winding and forming area, where it is cross-wound on the bobbin at a set winding angle. During this process, the system compares the operating parameters with the preset process tolerance value generated based on qualified batch statistics. If a yarn breakage, sudden drop in twist, abnormal bobbin ripples, or excessive forming deviation is detected, a process failure record is generated and the corresponding data segment is extracted.

[0016] Step S2: Based on the failed twisting sample, deduce the yarn twist unevenness to obtain yarn twist unevenness data; based on the yarn twist unevenness data, perform adaptive tension control adjustment of the yarn and output tension control adjustment data; In one implementation of this invention, when deriving yarn twist unevenness based on failed twisting sampling samples, the tension fluctuation data is first analyzed for time-domain fluctuation intensity. Specifically, each 20-second sampling data set is divided into 40 time periods of 0.5 seconds each, and the tension change amplitude for each time period is calculated to form a tension fluctuation segment sequence. Then, abnormal intervals are screened by combining the spindle tension and spindle speed data for the corresponding time periods. It should be noted that the tension reference threshold is obtained by collecting 30 consecutive sets of normal production data, and its calculation method is: T_threshold = T_average + 2σ, where T_threshold represents the tension reference threshold, T_average represents the average tension change amplitude under normal operating conditions, and σ represents the standard deviation. For example, if the average tension change amplitude under normal operating conditions is statistically obtained to be 0.18 cN / dtex and the standard deviation is 0.03 cN / dtex, then the tension reference threshold is set to 0.24 cN / dtex. If the change amplitude reaches 0.31 cN / dtex in a certain time period, it is determined to enter the abnormal set. Subsequently, the variance of tension fluctuations in the abnormal set is calculated, and the degree of imbalance in the winding tension distribution is calculated in conjunction with the spindle tension and spindle speed. For example, in a certain abnormal interval, when the spindle speed is 14,500 rpm and the spindle tension coefficient is 0.83, the variance of the spatial orientation distribution of tension is obtained as 0.027 through discrete analysis, from which the local twist imbalance index of the yarn is derived to be 0.41. Specifically, when the twist imbalance index exceeds 0.35, the system determines that the yarn has entered an unstable twisting state and initiates adaptive tension control adjustment. The control center reads the corresponding process parameters and dynamically corrects the traveler radius and air ring height in conjunction with the yarn processing fineness. For example, for 14.6 dtex yarn, the traveler radius is adjusted from 21 mm to 19 mm, and the air ring height is reduced from 205 mm to 192 mm to suppress centrifugal force fluctuations. Subsequently, based on the optimized process conditions, the tension intervention force was recalculated, increasing the initial tension control parameter from 0.07 cN / dtex to 0.12 cN / dtex. Simultaneously, the yarn contact point angle was corrected to 17° and the yarn tilt angle to 9°, ultimately outputting the tension control adjustment data. It is important to note that this adjustment process uses a step-by-step increment, with each adjustment not exceeding 0.01 cN / dtex, to avoid sudden changes in control that could lead to new yarn breakage risks.

[0017] Step S3: Optimize tension control adjustment data, then encode and package the data to output graded control coded data packets, and send them to the control center of the twisting machine to execute yarn running control.

[0018] In one implementation of this invention, the process of optimizing tension control adjustment data includes virtual operation simulation, boundary verification, and hierarchical encapsulation of control parameters.

[0019] Specifically, the control center invokes the equipment's digital twin operation module, imports the tension control adjustment data obtained in step S2 into the simulation environment, and establishes test conditions based on the actual equipment inertia parameters, spindle speed response delay, and yarn path friction coefficient. It then iteratively simulates the winding process over the next 60 seconds. It should be noted that the basic parameters of the digital twin operation module are obtained through on-site calibration; for example, the spindle response delay is 18 milliseconds, and the friction correction coefficient is 0.91, all obtained through backfitting from multiple start-stop tests on the actual equipment. During the simulation, the peak tension fluctuation, air ring amplitude, and yarn breakage prediction probability are monitored in real time. If a set of control data causes the peak tension fluctuation to exceed 0.26 cN / dtex, it is considered out of bounds, and the system automatically reverts to the first-level control parameters for re-optimization. Specifically, when a simulation shows a peak tension of 0.29 cN / dtex, the system reduces the tension intervention force from 0.13 cN / dtex to 0.11 cN / dtex, simultaneously lowering the yarn contact point angle by 1°, and re-executes the simulation verification. After successful verification, test condition verification data is generated. The verified control data is then subjected to hierarchical encoding, mapping the tension level, wire travel radius correction value, air travel height correction value, and angle correction parameter to control byte segments, and packaged according to control priority. Specifically, the tension level is mapped to the first byte, the air travel height to the second byte, and the angle control to the last check field, forming a complete hierarchical control coded data packet. It should be noted that a loop check field is added to the data packet before transmission to verify transmission integrity. After encapsulation, the data is sent to the twisting machine control center via the industrial bus. The control center parses the data and directly drives the tension actuator, wire travel guide adjustment mechanism, and winding angle adjustment mechanism to operate synchronously, thereby completing the closed-loop execution of yarn operation control.

[0020] The sampling of the recorded samples includes: The failure records of the yarn doubling process are timestamped, and then the failure records on adjacent timelines are removed as pre-processed failure records; A cross coordinate system of time and tension fluctuation is constructed based on the tension fluctuation data corresponding to the preprocessing failure records; Determine the asymmetric self-similar structure of tension fluctuations in a cross coordinate system; The difference in the rate of increase of tension fluctuations in the cross coordinate system is calculated based on the asymmetric self-similar structure. Based on the difference in the rate of increase of tension fluctuation, a sample of the preprocessing failure records is obtained to obtain the double twist failure sample.

[0021] In one implementation of this invention, after timestamp calibration and deduplication of adjacent failure records for preprocessing failure records, the control center cache module is first invoked to read the corresponding tension fluctuation data. The tension sampling points are then reconstructed and arranged according to a unified time reference to establish a cross-coordinate system of time and tension fluctuation. Specifically, the discrete tension values ​​collected by the tension sensor are mapped onto a two-dimensional analysis plane according to the sampling time sequence. The horizontal axis is defined as the running time axis, and the vertical axis is defined as the tension fluctuation amplitude axis. The origin of the time axis is selected as the position 15 seconds before the occurrence of each failure event, and the zero point of the vertical axis uses the average tension value within the corresponding time window as the reference center value. It should be noted that this average tension value is not directly taken from the original average value, but is obtained by detrending all tension data within a 20-second window. Specifically, linear drift caused by steady-state acceleration of the equipment is first removed, and then an average calculation is performed. For example, if a group of samples obtains an average tension value of 0.21 cN / dtex after detrending, this value is used as the zero reference reference for the vertical axis.

[0022] In one implementation of this invention, after constructing the cross coordinate system, a multi-scale interval segmentation analysis is performed on the tension evolution trajectory to identify whether an asymmetric self-similar structure exists. Specifically, the 20-second tension trajectory is divided into four observation windows of 8 seconds, 4 seconds, 2 seconds, and 1 second according to the time scale, and the distribution of local peaks, troughs, and adjacent inflection points in each observation window is extracted to establish a local tension morphology topology map.

[0023] Specifically, the system uses a left-right mirror difference analysis method to determine structural asymmetry. First, using each local peak point as the center, equal-length tension curves on the left and right sides of the peak are extracted to form mirror comparison sections. Then, the area difference between the two sides is calculated. If the area difference exceeds a preset asymmetry threshold for three consecutive scale windows, a significant asymmetric structure is determined to exist. It should be noted that the preset asymmetry threshold is calculated by adding 1.8 times the standard deviation to the average mirror area difference under historical normal operating conditions. For example, if the average is 0.043 and the standard deviation is 0.011, the preset asymmetry threshold is set to 0.063.

[0024] When determining self-similarity, morphological reconstruction is further performed on the tension trajectories at each scale window. Curves at different scales are proportionally compressed and mapped, and the consistency of peak and valley alignment after compression is compared. When a set of samples exhibits a peak-valley period repetition rate exceeding 82% at both the 4-second and 2-second scales, and a mirror difference value reaching 0.071, it is determined to have formed an asymmetric self-similar structure. It is important to note that this structure typically characterizes an abnormal coupling between spindle tension disturbance and local yarn torque accumulation, serving as a crucial basis for subsequently judging the evolution trend of tension instability. Therefore, once successfully identified, a corresponding asymmetric structure identifier is generated and written into the preprocessing failure record.

[0025] In one implementation of this invention, after identifying an asymmetric self-similar structure, a piecewise slope analysis is performed on the tension change process within the structure to calculate the difference in the rate of increase of tension fluctuations. Specifically, each asymmetric structure is divided into an ascending phase and a descending phase, and the tension change rate per unit time within each phase is extracted. The degree of tension instability propagation is assessed by the difference in rates between the two phases. Specifically, the calculation of the difference in the rate of increase of tension fluctuations is completed by subtracting the change rates of adjacent sampling segments. For example, in an abnormal structure, during the ascending phase, the tension increases from 0.19 cN / dtex to 0.33 cN / dtex in 0.42 seconds, so the ascending rate is 0.33 cN / dtex per second; during the descending phase, the tension decreases from 0.33 cN / dtex to 0.24 cN / dtex in 0.58 seconds, so the descending rate is 0.16 cN / dtex per second. The calculated difference in the rate of increase of tension fluctuations is 0.17 cN / dtex per second. It is important to note that if the growth rate difference of a certain structure exceeds the preset growth rate difference of 0.12 cN / dtex per second, the structure is considered to have a significant tendency for instability and diffusion. The preset growth rate difference is obtained by analyzing the statistical data of yarn breakage failure samples over 30 consecutive days, and its value corresponds to the 80th percentile of all abnormal samples. After completing the calculation of all structures, the growth rate differences are arranged in chronological order to form a tension fluctuation growth rate difference sequence, and the corresponding spindle speed and tension status are recorded simultaneously to form the basis for subsequent sampling.

[0026] In one implementation of this invention, based on the tension fluctuation growth rate difference sequence generated in step S13, optimal sampling is performed on the preprocessed failure records to extract the most representative twist instability segments. Specifically, the control center first sorts all failure records from high to low according to the tension fluctuation growth rate difference, and establishes a risk level stratification pool based on the fault type label.

[0027] Specifically, the growth rate difference was divided into three intervals: 0.12 to 0.18 cN / dtex per second was defined as a mild instability interval, 0.18 to 0.25 cN / dtex per second as a moderate instability interval, and greater than 0.25 cN / dtex per second as a severe instability interval. Samples were then drawn from each interval according to the principle of time-discrete distribution. It should be noted that the time-discrete distribution principle means that the time interval between adjacent sampling records must not be less than 120 seconds to avoid the same continuous failure process being repeatedly selected.

[0028] In one specific implementation, if a batch contains 126 preprocessing failure records, including 52 records of mild instability, 47 records of moderate instability, and 27 records of severe instability, then 15, 20, and 15 samples are drawn respectively in a 3:4:3 ratio, forming a total of 50 sets of double-twist failure sampling samples. For severe instability samples, the spindle speed change trajectory within 3 seconds before the failure occurs needs to be additionally added to enhance the accuracy of subsequent twist unevenness derivation.

[0029] It is important to note that after sampling, a consistency check is required. If more than 8% of the sampling points are missing within a sample window, they will be automatically removed and supplemented from the candidate records of the same level. The final failed twisting sampling samples are uniformly written into the analysis buffer and directly associated with the fluctuation intensity analysis module in step S2 to achieve a closed-loop data connection for the derivation of yarn twist unevenness.

[0030] refer to Figure 2 As shown, step S2 includes: S201: Analyze the fluctuation intensity of tension fluctuation data in the time dimension of the failed twisting sample; In one implementation of this invention, the time-dimensional fluctuation intensity of the failed twisting samples generated in step S14 is analyzed. Specifically, each 20-second tension fluctuation data set is divided into 40 analysis time periods of 0.5 seconds each, and the tension change amplitude of each time period is calculated to form a segmented sequence of fluctuation intensity. The tension change amplitude is determined by the difference between the maximum and minimum values ​​of the segment, expressed as: Q = Tmax - Tmin, where the fluctuation intensity Q represents the tension change amplitude of the current time period, the maximum tension value Tmax represents the maximum value of the sampling point within that time period, and the minimum tension value Tmin represents the minimum value. It should be noted that, to ensure calculation stability, both the maximum and minimum values ​​are statistical results after removing the first and last two abnormal sampling points. For example, if the effective tension value range for a certain time period is 0.17 cN / dtex to 0.31 cN / dtex, then the fluctuation intensity is 0.14 cN / dtex. Subsequently, the fluctuation intensity over all time periods is compared with the normal operating condition threshold. This normal operating condition threshold is obtained through statistical analysis of normal winding data; for example, the mean of 0.16 cN / dtex is added to twice the standard deviation of 0.04 cN / dtex to obtain 0.24 cN / dtex. If a certain time period exceeds this value, it is marked as an abnormal fluctuation range, and fluctuation intensity analysis data is generated.

[0031] S202: Based on the fluctuation intensity and the sampling of failed twisting, the yarn twist unevenness is deduced to obtain the yarn twist unevenness data; In one implementation of this invention, the yarn twist imbalance state is deduced based on the fluctuation intensity analysis data generated in step S201, combined with the spindle tension and spindle speed information in the sampled samples. Specifically, all abnormal fluctuation intervals are first screened, and variance calculation is performed on the tension fluctuations in the abnormal intervals to reflect the local tension dispersion; then, the winding tension distribution imbalance coefficient is derived by combining the spindle speed and spindle tension at the corresponding time. It should be noted that the spindle tension is measured by the displacement sensor of the spindle tensioning device, and the spindle speed is obtained by the spindle speed encoder and converted into a normalized value between 0 and 1.

[0032] Specifically, the system calculates the twist unevenness index based on the winding tension distribution imbalance coefficient. When this index exceeds 0.35, the yarn is determined to have significant twist unevenness. For example, in an abnormal interval, the tension variance is 0.026, the spindle speed is 14,500 rpm, and the spindle belt tension is 0.82. After coupling and conversion, the twist unevenness index is calculated to be 0.41, thus indicating that the yarn has entered an unstable twisting state. It should be noted that, to avoid misjudgment caused by instantaneous mechanical impact, valid yarn twist unevenness data is only generated and sent to the molding process optimization module when the twist unevenness index exceeds the limit for three consecutive analysis time periods.

[0033] S203: Optimize molding process conditions based on yarn twist unevenness data and fluctuation intensity to obtain optimized process condition data; In one implementation of this invention, after receiving yarn twist unevenness data, the control center optimizes the forming process parameters based on the current yarn processing fineness and fluctuation intensity. Specifically, it first calls the process database to read the current processing yarn fineness, for example, 14.6 dtex; then, it matches the steel traveler running radius adjustment range according to the preset process adaptation table (here, the preset process adaptation table refers to the process processing requirements table that the manufacturer needs to design for the yarn, which is input to the control center during processing).

[0034] Specifically, when the twist imbalance index is between 0.35 and 0.45, the traveler radius is reduced by 1mm to 3mm; when it exceeds 0.45, the balloon height is adjusted simultaneously. It should be noted that the balloon height adjustment range is limited to 180mm to 220mm to avoid yarn collision with the yarn guide structure. For example, if a sample has a twist imbalance index of 0.41 and a peak fluctuation intensity of 0.29cN / dtex, the traveler radius is adjusted from 21mm to 19mm, and the balloon height is adjusted from 205mm to 193mm. After adjustment, the centrifugal force distribution is re-simulated. If the centrifugal force imbalance coefficient decreases by more than 15%, the optimization is confirmed as effective, and optimized process condition data is output, including traveler correction values, balloon height correction values, and suggested winding angle correction values.

[0035] S204: Based on yarn twist mismatch data and process condition optimization data, perform adaptive tension control adjustment of yarn and output tension control adjustment data.

[0036] In one implementation of this invention, the twisting machine undergoes adaptive tension parameter adjustment based on yarn twist misalignment data and optimized process conditions. Specifically, the system first calls the current initial tension control parameters, typically set to 0.06 cN / dtex to 0.08 cN / dtex; then, it derives the tension intervention force by combining the optimized traveler radius and balloon height.

[0037] Specifically, the tension adjustment adopts a step-by-step incremental strategy, with each increment controlled within 0.01 cN / dtex, while simultaneously correcting the yarn contact point angle and tilt angle. It should be noted that the angle correction range is based on preset mechanical limits of the equipment, with the contact point angle limited to between 15° and 18°, and the tilt angle limited to between 7° and 10°. For example, if the initial tension control value is 0.07 cN / dtex, and the target intervention force is derived from the optimized data to be 0.12 cN / dtex, then the adjustment is gradually increased in 5 levels, while simultaneously correcting the yarn contact point angle from 15° to 17° and the tilt angle from 8° to 9°. After each level of adjustment, real-time tension feedback is monitored. If the tension fluctuation is below 0.22 cN / dtex for two consecutive seconds, the adjustment is confirmed as effective, and finally, tension control adjustment data is generated and sent to step S3 for verification and optimization.

[0038] The derivation of the yarn twist unevenness includes: Calculate the amplitude of fluctuation intensity change for each time period; obtain the spindle tension and spindle speed for the corresponding time period from the failed twisting sample; The abnormal set of variation amplitude is filtered out based on the preset tension reference threshold; the abnormal set refers to the part of the periodic variation amplitude that is greater than the tension reference threshold. Estimate the variance of abnormal tension changes in the abnormal set; Based on the variance of abnormal tension changes, the tightness of the spindle belt, and the spindle speed, the unbalance of the winding tension distribution of the yarn is calculated, and the unbalance of the winding tension distribution is obtained. Based on the unbalanced winding tension distribution data, the yarn twist unevenness data is derived.

[0039] In one implementation of this invention, the amplitude of each segment change in the fluctuation intensity segment sequence formed in step S201 is extracted, and the corresponding spindle tension and spindle speed parameters are simultaneously matched. Specifically, each group of 20-second double-twist failure sampling samples is further divided into 40 analysis segments with a time resolution of 0.5 seconds. The peak and valley values ​​of the tension fluctuation curve are extracted within each analysis segment, and the amplitude of change is calculated. The expression for calculating the amplitude of change is: B = Tfeng - Tgu, where the amplitude of change B represents the amplitude of tension fluctuation in the current time segment, Tfeng represents the maximum sampled tension value within the segment, and Tgu represents the minimum sampled tension value. It should be noted that both the peak and valley values ​​are extracted using 5-point smoothing to avoid local noise disturbances. For example, if the peak tension value of a certain segment is 0.34 cN / dtex and the valley value is 0.18 cN / dtex, then the amplitude of change is 0.16 cN / dtex. Specifically, the control center simultaneously retrieves the spindle tension and spindle speed information for the corresponding time segment. The tension of the spindle belt is obtained through conversion using a tensioning wheel displacement detector. A standard loading test is used to establish a mapping relationship between displacement and tension coefficient; for example, when the detected displacement is 0.72 mm, the tension coefficient is converted to 0.81. The spindle speed is acquired in real time by the spindle encoder; for example, the spindle speed in the corresponding section is 14620 rpm. Finally, joint analysis data including variation amplitude, spindle belt tension, and spindle speed are generated, providing an input basis for anomaly screening.

[0040] In one implementation of this invention, anomaly set filtering is performed based on the joint analysis data of the variation amplitude. Specifically, a tension reference threshold is first set, which is obtained by statistically analyzing 10 sets of tension data under normal winding conditions continuously collected from historical records. It should be noted that the normal winding condition samples must meet the requirements of no yarn breakage, no winding offset exceeding the limit, and stable tension fluctuation for 30 consecutive minutes. The tension reference threshold is established by calculating the mean and standard deviation of the variation amplitudes of all normal samples.

[0041] Specifically, if the statistically obtained normal variation amplitude is 0.17 cN / dtex and the standard deviation is 0.035 cN / dtex, then the tension reference threshold is set to 0.24 cN / dtex. The system compares the target variation amplitude with the tension reference threshold segment by segment and further determines whether it has periodic abnormal characteristics. Periodicity is determined by repeated exceedances in three or more consecutive analysis segments. For example, if the variation amplitudes in segments 12 to 15 are 0.26, 0.29, 0.28, and 0.27 cN / dtex respectively, then a periodic anomaly is determined to have occurred.

[0042] It is important to note that isolated single-point exceedances are not included in the anomaly set to prevent accidental triggering by sporadic mechanical vibrations. After screening, all analysis segments that meet the continuous cycle exceedance criteria are combined to generate an anomaly set, and their start and end times and corresponding spindle speed ranges are recorded. For example, if an anomaly set covers the period from the 8th to the 10th second, corresponding to a spindle speed fluctuation range of 14,500 to 14,720 revolutions per minute, it is written into the anomaly segment index table for subsequent estimation of the variance of abnormal tension changes.

[0043] In one implementation of this invention, the dispersion of tension changes within the anomaly set is estimated to quantify the local tension instability propagation trend. Specifically, the statistical mean of all variation amplitude samples within the anomaly set is calculated, and the deviation of each sample from the mean is calculated to form the variance of the anomaly tension changes. Its expression is: ;in, The variance of abnormal tension changes indicates the degree of dispersion of the abnormal set. The sample change amplitude represents the change amplitude of the i-th analysis segment. The average change magnitude represents the average value within the outlier set. The sample size represents the total number of abnormal segments. For example, if an abnormal set contains four segments with variation amplitudes of 0.26, 0.29, 0.31, and 0.28 cN / dtex, the average variation amplitude is 0.285 cN / dtex, and the calculated variance of the abnormal tension variation is 0.00033. It should be noted that to enhance the sensitivity of the results, a trend correction should be performed on the abnormal segments before calculation to remove systematic drift caused by the overall speed increase of the equipment.

[0044] Specifically, when the variance of abnormal tension changes exceeds 0.00025, the system determines that there is significant tension disturbance propagation. If it exceeds 0.00040, it is marked as a strong anomaly. It should be noted that the variance result reflects not only the magnitude of the tension fluctuation but also the degree of instability. Therefore, it is used as an important input parameter for calculating winding tension imbalance and participates in the coupled analysis together with the spindle tension and spindle speed.

[0045] In one implementation of this invention, the control center performs a winding tension distribution imbalance calculation based on the abnormal tension change variance, the spindle tension, and the spindle speed. Specifically, firstly, the offset trend of the wire coil rotation trajectory is derived based on the spindle tension and the spindle speed, and then this offset trend is coupled with the abnormal tension change variance to generate the tension spatial orientation distribution state.

[0046] It should be noted that the trend of wire coil trajectory deviation is obtained through conversion from the equipment calibration database (i.e., a database compiled from previous records of wire coil rotation trajectory deviation, which can be retrieved from the control center of the twisting machine). For example, when the spindle tension coefficient is below 0.80 and the spindle speed is above 14,500 rpm, the trajectory deviation increase is approximately 18%. If the current abnormal tension variation variance is 0.00033, the spindle tension coefficient is 0.79, and the spindle speed is 14,680 rpm, then the winding trajectory is determined to have an eccentric instability trend. Specifically, the winding tension distribution imbalance index is calculated by interpolation through a lookup table. For example, the imbalance index is 0.63 under the corresponding parameter combination. It should be noted that the lookup table data comes from the calibration of 30 sets of equipment operation tests in the early stage, and each set of tests records the relationship between the spatial distribution of tension and the forming quality deviation. Finally, the system encapsulates the imbalance index and its corresponding time interval to form winding tension distribution imbalance data and transmits it to the twist unevenness derivation module.

[0047] In one implementation of this invention, the yarn twist imbalance state is derived based on the unbalanced winding tension distribution data. Specifically, the system first establishes a mapping relationship between the imbalance index and the twist deviation. This mapping relationship is obtained through regression fitting of 20 historical sets of double-twist process detection results. The detection method uses an offline twist meter to measure the actual number of twists per unit length of the yarn.

[0048] Specifically, when the winding tension distribution imbalance index is in the range of 0.40 to 0.55, it corresponds to mild twist imbalance; when it is in the range of 0.55 to 0.70, it corresponds to moderate imbalance; and when it exceeds 0.70, it corresponds to severe imbalance. For example, if the winding tension distribution imbalance index of a certain group is 0.63, then the twist imbalance index is mapped to 0.41, which is judged as a moderate imbalance state.

[0049] It should be noted that, to avoid a single anomaly affecting control stability, the system needs to perform a moving average correction on five consecutive sets of imbalance data. When the corrected twist misalignment index exceeds 0.35 for three consecutive times, the yarn twist misalignment data is officially generated, and a misalignment level label and recommended control level are attached. The final yarn twist misalignment data is used for subsequent optimization of molding process conditions and adaptive tension adjustment.

[0050] The abnormal set refers to the portion of the change amplitude in each time period that exceeds the tension reference threshold.

[0051] The calculation of yarn winding tension distribution imbalance includes: The concentricity center alignment offset is calculated based on the spindle belt tension and spindle speed. The discrete fluctuation of the tensile force of the spindle strip during yarn winding is estimated based on the alignment offset. The variance of the spatial orientation distribution of yarn tension during the winding process is calculated based on the discrete fluctuation of tensile force and the variance of abnormal tension changes. The unbalance of the winding tension distribution of yarn is calculated based on the variance of the spatial orientation distribution of tension, and the unbalanced winding tension distribution data is obtained.

[0052] In one implementation of this invention, the control center calculates the concentricity center alignment offset based on the spindle tension and spindle speed data extracted in step S202 to assess the degree of deviation of the wire coil's running trajectory relative to the theoretical winding center. Specifically, firstly, the initial calibration parameters of the equipment are read. These parameters are obtained by continuously collecting 50 sets of standard trajectory data under no-load operation to establish the theoretical concentricity center reference coordinates. Subsequently, the real-time spindle tension and spindle speed are substituted into the offset conversion model to calculate the actual running center offset value. The expression of the offset conversion model is as follows: Among them, the position offset This represents the offset value of the wire coil's running center, and the correction factor. Indicates the equipment structure calibration coefficient, tightness This indicates the real-time normalized tension coefficient of the spindle belt and the real-time spindle speed. Indicates the current spindle speed, reference spindle speed This indicates the rated stable spindle speed of the equipment. It should be noted that the correction factor is determined through on-site calibration, and is typically set to 0.52 after continuous testing of the equipment.

[0053] Specifically, if the current spindle tension coefficient is 0.79, the real-time spindle speed is 14,600 rpm, and the reference spindle speed is 14,000 rpm, then the calculated alignment offset is approximately 0.11 mm. It is important to note that when the offset exceeds 0.08 mm, a significant risk of trajectory deviation is identified, and the corresponding offset data is written to the winding state buffer to provide an input basis for subsequent estimation of discrete fluctuations in tensile force.

[0054] In one implementation of this invention, the discrete fluctuation state of the tension applied to the yarn by the spindle is estimated based on the concentricity center alignment offset. Specifically, the control center calls the equipment dynamic coupling database, which collects 30 sets of winding data under different offset conditions from historical data, and performs data clustering and summarization based on the winding data to record the relationship between the offset and the change in tension force.

[0055] Specifically, the system obtains the corresponding discrete amplitude of the pulling force based on the equipment's dynamic coupling database and combines it with real-time spindle speed correction. For example, when the alignment offset is in the range of 0.10mm to 0.12mm, the corresponding basic pulling force fluctuation value is 0.045N under the condition of a spindle speed of 14500 rpm; if the current spindle speed increases to 14600 rpm, an additional correction of 0.002N is added for every 100 rpm increase, finally obtaining a discrete fluctuation value of 0.047N for the pulling force.

[0056] It should be noted that, to avoid the influence of short-term impact errors on the estimation results, the system performs a moving average processing on the tensile force data for five consecutive sampling periods. If the smoothed discrete fluctuation value exceeds 0.040N, it is marked as a significantly discrete state. Specifically, when the system reaches a significantly discrete state for three consecutive time windows, it generates tensile force discrete fluctuation data and synchronously transmits it to the tension spatial orientation distribution calculation with corresponding time labels.

[0057] In one implementation of this invention, the spatial orientation distribution variance of tension is calculated jointly based on the discrete fluctuation data of tensile force and the variance of abnormal tension changes to describe the degree of tension dispersion diffusion of the yarn in the winding space. Specifically, the discrete fluctuation of tensile force is regarded as a radial disturbance input, and the variance of abnormal tension changes is regarded as an axial disturbance input, and the two are coupled and analyzed.

[0058] The variance of the spatial orientation distribution of tension is calculated as follows:

[0059] Among them, the variance of spatial orientation distribution Indicates the degree of dispersion of tension in the winding space, and the variance of abnormal tension changes. This represents the discrete statistical value of tension obtained from the preceding steps, and the discrete fluctuation of the pulling force. Represents the radial pull disturbance value, coupling coefficient and correction factor The coefficients were obtained through experimental calibration. It should be noted that the two coefficients were established using a minimum error fitting method, for example, in this device, they were set to 0.62 and 0.38 respectively.

[0060] Specifically, if the variance of abnormal tension variation is 0.00033 and the discrete fluctuation of pulling force is 0.047 N, then the calculated variance of the spatial orientation distribution of tension is 0.018. It should be noted that when this value exceeds 0.015, it indicates a significant skewed diffusion of yarn tension in its spatial distribution. Based on this, the system generates tension spatial orientation distribution analysis data and records its duration and the corresponding winding layer position.

[0061] In one implementation of this invention, a winding tension distribution imbalance calculation is performed based on the variance of the spatial orientation distribution of tension. Specifically, the control center calls the winding state mapping library to map the variance of the spatial orientation distribution to the winding tension imbalance level. The winding state mapping library is established using 20 sets of winding quality inspection samples from the failed twisting sample collection, where each set of samples synchronously records spatial tension distribution data and finished yarn bobbin forming deviation.

[0062] Specifically, when the variance of the spatial orientation distribution of tension is between 0.010 and 0.015, it corresponds to a slight imbalance; when it is between 0.015 and 0.022, it corresponds to a moderate imbalance; and when it exceeds 0.022, it corresponds to a severe imbalance. For example, if the variance of the spatial orientation distribution of a sample is 0.018, then the resulting winding tension imbalance index is 0.63, and it is determined to be a moderate imbalance.

[0063] It should be noted that, to improve calculation stability, the system also incorporates the current winding layer thickness for correction. When the winding layer thickness exceeds 18mm, a 5% correction gain is added to the imbalance index to reflect the amplification effect of the offset under thick winding conditions. This ultimately generates winding tension distribution imbalance data, including the imbalance index, imbalance level, and duration range, which is then sent to the yarn twist unevenness derivation module to generate the final yarn twist unevenness data.

[0064] The optimization of molding process conditions includes: Obtain the actual processed fineness of the yarn; The traveler running radius is adjusted based on the yarn twist unevenness data and the actual processed fineness. The intensity of centrifugal force imbalance is determined by the intensity of fluctuation; the height of the air ring is adjusted based on the intensity of centrifugal force imbalance, yarn twist unevenness data and actual processed fineness, and the adjustment range is controlled between 180mm-220mm, and the air ring height adjustment data is output. The forming process conditions are optimized based on the operating radius and air ring height adjustment data of the wire ring to obtain optimized process condition data.

[0065] In one implementation of this invention, the control center first acquires the actual processed fineness of the current yarn as the basic input parameter for subsequent adjustment of the traveler's running radius and balloon height. Specifically, an online quality detection module is installed at the yarn guide inlet of the twisting machine. This module continuously collects the yarn mass per unit length and performs real-time calculations based on the running length to obtain the actual processed fineness of the yarn. It should be noted that the online quality detection module needs to be calibrated before the equipment is started. The calibration method involves using three sets of standard fineness samples for comparison calibration, corresponding to 12.0 dtex, 14.0 dtex, and 16.0 dtex respectively, to establish a mapping relationship between the quality signal and the fineness.

[0066] Specifically, the system generates a fineness statistical value every 30 seconds and takes the average of three consecutive test results as the current actual processed fineness. For example, if a test result is 14.4 dtex, 14.7 dtex, and 14.8 dtex, the average fineness is 14.63 dtex. It is important to note that when the consecutive test deviation exceeds 0.5 dtex, the system automatically determines that there is an abnormal yarn dryness. In this case, process condition optimization is temporarily suspended until the fineness fluctuation stabilizes, at which point the calculation is recalculated. Finally, the actual processed fineness of the yarn is generated.

[0067] In one implementation of this invention, the control center adjusts the traveler's running radius based on yarn twist unevenness data and the actual processed fineness to adjust the tension distribution during yarn winding. The optimal running radius under different fineness and twist unevenness conditions is recorded.

[0068] The operating radius of the wire coil is expressed using the following adaptive expression:

[0069] Among them, the adaptation radius Indicates the target wire coil's running radius, the reference radius. Indicates the standard radius of the equipment and the unevenness of the twist. The index represents the average loss data generated in the preceding steps, and the actual processed fineness. Indicates real-time fineness detection, standard fineness. Indicates the fineness of the process design, correction factor and compensation coefficient Obtained through experimental calibration. It should be noted that the correction and compensation coefficients were established through minimum error fitting, and were set to 4.2 and 0.6 respectively. If the current reference radius is 21mm, the twist misalignment index is 0.41, the actual fineness is 14.6 dtex, and the standard fineness is 14.0 dtex, then the calculated fitting radius is approximately 19mm. It is important to note that the adjustment range of the coil running radius is limited to between 18mm and 22mm to avoid exceeding the mechanical limits of the guide rail mechanism. The final coil running radius adjustment data is then generated.

[0070] In one implementation of this invention, the centrifugal force imbalance intensity is determined based on the fluctuation intensity analysis results obtained in step S201, and the balloon height is dynamically adjusted in conjunction with the yarn twist unevenness data and the actual processed fineness. Specifically, the system takes the average of the fluctuation intensity over 10 consecutive analysis time periods and compares it with the equipment calibration centrifugal force reference table.

[0071] It should be noted that the centrifugal force reference table was obtained through field tests under different spindle speeds and gas ring conditions. For example, when the average fluctuation intensity is in the range of 0.22 cN / dtex to 0.26 cN / dtex, it corresponds to mild centrifugal imbalance; when it is in the range of 0.26 cN / dtex to 0.32 cN / dtex, it corresponds to moderate centrifugal imbalance; and when it exceeds 0.32 cN / dtex, it corresponds to severe centrifugal imbalance.

[0072] Specifically, when a moderate centrifugal imbalance is determined, the gas ring height is corrected by combining the twist misalignment index and the actual fineness. For example, if the current average fluctuation intensity is 0.29 cN / dtex, the twist misalignment index is 0.41, and the actual fineness is 14.6 dtex, then the gas ring height will be adjusted from 205 mm to 193 mm. It is important to note that the adjustment is performed in incremental increments, with each adjustment controlled within 3 mm, and tension feedback is monitored in real time. If the peak tension decreases by more than 12% after adjustment, the correction is considered effective. Finally, the gas ring height adjustment data is generated and written to the control buffer.

[0073] In one implementation of this invention, the control center integrates the wire coil running radius adjustment data and the air ring height adjustment data to optimize the forming process conditions, thereby forming a complete process adjustment strategy. Specifically, a 30-second virtual winding simulation is performed on the adjusted parameter combination, and the tension peak, air ring amplitude, and winding trajectory deviation are monitored in real time.

[0074] It should be noted that the inertia parameters and friction correction parameters of the digital process simulation module are obtained through on-site equipment calibration. For example, the spindle inertia correction value is 1.08, and the yarn guide friction correction coefficient is 0.93. If the peak tension fluctuation during the simulation is less than 0.24 cN / dtex, and the winding trajectory offset is less than 0.05 mm, then the parameter combination is considered to meet the optimization requirements.

[0075] Specifically, for example, after adjusting the wire coil's running radius to 19mm and the air ring height to 193mm, simulations show a peak tension of 0.21cN / dtex and a trajectory offset of 0.038mm, confirming the validity of this parameter set. The system further generates process condition optimization data, including correction values ​​for the wire coil radius, air ring height, recommended winding angle, and corresponding priority execution levels. It is important to note that this optimization data will be directly transmitted to the tension adaptive adjustment module in step S204 to complete the closed-loop control connection.

[0076] The process of adaptive tension control adjustment of yarn includes: Set the initial tension control parameters, and control the initial tension between 0.06 and 0.08 CN / dtex; Based on yarn twist unevenness data and process condition optimization data, the yarn angle range was optimized and adjusted, including the yarn contact point angle range and the yarn tilt angle range. Based on the optimized process conditions and yarn angle range, the range of intervention force for tension control in the twisting zone is derived, and the range of intervention force for tension control in the twisting zone needs to be constrained to between 0.10-0.15 cN / dtex. Based on the tension intervention level data, the initial tension control parameters are adjusted for yarn adaptive tension control, and the tension control adjustment data is output.

[0077] In one implementation of this invention, the control center optimizes the yarn running angle range based on yarn twist unevenness data and process condition optimization data. Specifically, the system reads the steel traveler running radius correction value and air ring height adjustment data, and combines this with the twist unevenness index to deduce the yarn path geometry. The yarn contact point angle adjustment uses the following expression: Among them, the angle of the target contact point This indicates the angle of the yarn contact point after adjustment; the reference contact angle. Indicates the default angle of the equipment, twist unevenness index. This indicates the output value before the step, and the air sphere height correction amount. This represents the optimized height change value, and the correction coefficient. and compensation coefficient Obtained through equipment calibration. It should be noted that during equipment calibration, a correction factor of 3.5 and a compensation factor of 0.04 were determined through 20 sets of winding tests. Specifically, if the reference contact angle is 15°, the twist unevenness index is 0.41, and the loop height correction is 12mm, then the calculated contact point angle is approximately 17°. Simultaneously, the yarn tilt angle is corrected synchronously according to the traveler's running radius, for example, from 8° to 9°. It is important to note that the yarn contact point angle is limited to between 15° and 18°, and the tilt angle is limited to between 7° and 10°. Automatic truncation and correction are applied when these ranges are exceeded. The final result is optimized yarn angle range data.

[0078] In one implementation of this invention, the control center derives the range of tension control intervention force in the twisting zone based on process condition optimization data and yarn angle range. Specifically, the running radius of the steel traveler, the height of the air ring, and the yarn angle parameters are input into the tension intervention force derivation. A parameter mapping relationship library is constructed through training with 30 historical stable winding samples.

[0079] It should be noted that the system prioritizes estimating the additional tension requirement based on the degree of angular deviation. For example, for every 1° increase in the yarn contact point angle, the additional tension requirement increases by approximately 0.008 cN / dtex; for every 1° increase in the tilt angle, the additional tension requirement increases by 0.005 cN / dtex.

[0080] Specifically, if the current contact point angle is adjusted from 15° to 17° and the tilt angle from 8° to 9°, the theoretical additional intervention force will increase by approximately 0.021 cN / dtex. If the initial tension control value is 0.07 cN / dtex, the derived target tension intervention force is 0.12 cN / dtex. It is important to note that the derived result must be limited to between 0.10 cN / dtex and 0.15 cN / dtex; if it is below the lower limit, it will be executed at 0.10 cN / dtex; if it is above the upper limit, it will be limited to 0.15 cN / dtex to prevent excessive pressure from causing yarn breakage. The final tension intervention force range data is then generated and transmitted to the adjustment execution module.

[0081] In one implementation of this invention, the control center performs a step-by-step adaptive adjustment of the initial tension control parameters based on the tension intervention force range data. Specifically, a phased incremental control strategy is adopted, with a single-stage adjustment step size of 0.01 cN / dtex, to gradually increase the initial tension to the target intervention force.

[0082] It should be noted that after each adjustment stage, the system maintains the current tension state for 2 seconds and collects tension feedback data in real time to determine whether the tension fluctuation has entered the stable range. The stable range is determined by statistical analysis of successfully wound samples in the manufacturer's database, and its upper limit is 0.22 cN / dtex.

[0083] Specifically, for example, if the initial tension control value is 0.07 cN / dtex and the target intervention force is 0.12 cN / dtex, then the force is increased in five increments: 0.08 cN / dtex, 0.09 cN / dtex, 0.10 cN / dtex, 0.11 cN / dtex, and 0.12 cN / dtex. When adjusting to 0.11 cN / dtex, if the tension fluctuation has decreased to 0.20 cN / dtex for two consecutive seconds, it is considered that a stable condition has been reached, and the increment can be stopped early; otherwise, the process continues until the target value is reached.

[0084] It is important to note that if a tension peak exceeding 0.26 cN / dtex is detected after any level adjustment, the system immediately reverts to the previous level of parameters and re-verifies. The final confirmed control value, together with the corresponding angle correction, generates tension control adjustment data and sends it to the simulation verification module in step S3 for closed-loop optimization.

[0085] Step S3 includes: Iterative simulation tests were conducted on the tension control adjustment data, and test data were collected. Perform boundary condition verification based on test data and output test condition verification data; Based on the test conditions, the data is verified and the tension control adjustment data is optimized. Then, the data is encoded and packaged to output a graded control coded data packet, which is sent to the control center of the twisting machine to execute yarn operation control.

[0086] In one implementation of this invention, after receiving the tension control adjustment data generated in step S204, the control center first performs an iterative simulation test to verify the adaptability of the adjustment parameters in the actual winding environment. Specifically, the control center calls the digital twin simulation module to import the tension control value, the running radius of the wire coil, the height of the air ring, and the yarn angle correction parameters into the digital twin simulation module. The module is established by collecting parameters such as equipment inertia, friction coefficient, and response delay on-site. Its basic parameters are obtained through backfitting of 50 consecutive start-stop tests; for example, the spindle response delay is calibrated to 18ms, and the friction correction coefficient is calibrated to 0.93.

[0087] Specifically, the system performs 10 iterative simulations on the target parameters in 60-second cycles. Each simulation cycle records the peak tension fluctuation, winding trajectory offset, and yarn breakage prediction probability. It should be noted that when the peak tension reaches 0.27 cN / dtex and the trajectory offset reaches 0.06 mm in a certain test cycle, it is recorded as an abnormal test node. After all tests are completed, a test data set is generated, including the operational stability indicators for each cycle, the distribution of abnormal nodes, and parameter response curves, and written to the test buffer for subsequent boundary verification.

[0088] In one implementation of this invention, the control center performs boundary condition verification based on the test data set to determine whether the tension control adjustment parameters meet the constraints for safe operation of the equipment. Specifically, the system calls the boundary condition determination module to compare each key indicator in the test data with preset boundary values.

[0089] Boundary verification uses the following decision expression:

[0090] Among them, the comprehensive verification coefficient Indicates the overall boundary fit, peak tension. This represents the maximum tension value and tension boundary value in the simulation test. The trajectory offset is set to 0.26 cN / dtex. This represents the maximum offset value of the winding center, and the offset boundary value. Set to 0.05mm, yarn breakage prediction probability This represents the risk value of abnormal yarn breakage, a probability boundary value. Set to 0.08.

[0091] It should be noted that the preset boundary values ​​were obtained by statistically analyzing the extreme operating data of stable production batches over the past 6 months. Specifically, verification is considered passed when the comprehensive verification coefficient is less than or equal to 3; if it exceeds 3, it indicates that at least one key indicator has exceeded the limit. For example, in a certain test group, the peak tension was 0.25 cN / dtex, the trajectory offset was 0.048 mm, and the yarn breakage prediction probability was 0.06, so the comprehensive verification coefficient was approximately 2.66, which was considered to meet the boundary requirements.

[0092] It is important to note that if two consecutive rounds of testing fail, the current parameter combination will be automatically marked as a failure group. The system will then generate test condition verification data, including verification status, out-of-bounds indicator types, and recommended correction directions.

[0093] In one implementation of this invention, the control center performs secondary optimization of the tension control adjustment parameters based on test condition verification data. Specifically, when there are out-of-bounds indicators in the test condition verification data, the system corrects the tension control value, yarn contact point angle, and air ring height in order of priority. For example, if the tension peak exceeds the limit, the tension control value is reduced by 0.01 cN / dtex; if the trajectory deviation exceeds the limit, the contact point angle is reduced by 1° simultaneously. After optimization, a rapid simulation verification is performed again, and after confirmation, the system enters the coding and encapsulation stage.

[0094] It should be noted that the control parameter coding adopts a hierarchical mapping method, mapping the tension control value to the first control field, the wire coil running radius to the second field, the air ring height to the third field, and the angle correction value to the verification field. For example, the tension control value of 0.11cN / dtex is mapped to control level "4", the wire coil radius of 19mm is mapped to "2", and the air ring height of 193mm is mapped to "5".

[0095] Specifically, the system concatenates fields according to control priority and adds a cyclic check field to generate a complete hierarchical control code data packet. This packet is then sent to the twisting machine control center via the industrial bus. Upon receiving the packet, the control center parses it and drives the tension actuator, wire guide adjustment mechanism, and winding angle correction mechanism to operate synchronously. It is important to note that after sending the data packet, a confirmation signal from the equipment is required. If no confirmation is received within 100ms, the packet is automatically retransmitted. This ultimately achieves closed-loop execution of yarn operation control.

[0096] The present invention also provides a yarn running control system for executing the yarn running control method described above, the yarn running control system comprising: The sample sampling module is used to extract yarn twisting process failure records from the control center of the twisting machine. The yarn twisting process failure records include tension fluctuation data obtained from the tension sensor. The module performs record sampling on the yarn twisting process failure records to obtain twisting failure sampling samples. The graded control module is used to deduce the yarn twist unevenness based on the failed twisting sample, and obtain the yarn twist unevenness data; based on the yarn twist unevenness data, it performs adaptive tension control adjustment of the yarn and outputs tension control adjustment data; The encoding module is used to optimize tension control adjustment data, and then encode and package the data to output graded control encoding data packets, which are then sent to the control center of the twisting machine to execute yarn running control.

[0097] Therefore, the embodiments should be considered as exemplary and non-limiting in all respects, and the scope of the invention is defined by the appended claims rather than the foregoing description. Thus, all variations falling within the meaning and scope of the equivalents of the application are intended to be included within the invention.

[0098] The above description is merely a specific embodiment of the present invention, enabling those skilled in the art to understand or implement the invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the general principles defined herein may be implemented in other embodiments without departing from the spirit or scope of the invention. Therefore, the present invention is not to be limited to the embodiments shown herein, but is to be accorded the widest scope consistent with the principles and novel features of the invention herein.

Claims

1. A yarn running control method, characterized in that, Applied to a doubling machine, a tension sensor is deployed in the contact area between the yarn and the bobbin, including the following steps: Step S1: Extract yarn twisting process failure records from the control center of the twisting machine. The yarn twisting process failure records include tension fluctuation data obtained from the tension sensor. Sample the yarn twisting process failure records to obtain a twisting failure sampling sample. Step S2: Based on the failed twisting sample, deduce the yarn twist unevenness to obtain yarn twist unevenness data; based on the yarn twist unevenness data, perform adaptive tension control adjustment of the yarn and output tension control adjustment data; Step S3: Optimize tension control adjustment data, then encode and package the data to output graded control coded data packets, and send them to the control center of the twisting machine to execute yarn running control.

2. The yarn running control method according to claim 1, characterized in that, Step S2 includes: Analyze the fluctuation intensity of tension fluctuation data in the time dimension of the failed twisting sample; Based on the fluctuation intensity and the sampling of failed twisting, the yarn twist unevenness was deduced, and the yarn twist unevenness data was obtained. Based on the yarn twist unevenness data and fluctuation intensity, the molding process conditions were optimized to obtain process condition optimization data; Based on yarn twist unevenness data and process condition optimization data, adaptive tension control adjustment of yarn is performed, and tension control adjustment data is output.

3. The yarn running control method according to claim 2, characterized in that, The derivation of the yarn twist unevenness includes: Calculate the amplitude of fluctuation intensity change for each time period; obtain the spindle tension and spindle speed for the corresponding time period from the failed twisting sample; The set of abnormal values ​​with varying amplitudes is selected based on a preset tension reference threshold. Estimate the variance of abnormal tension changes in the abnormal set; Based on the variance of abnormal tension changes, the tightness of the spindle belt, and the spindle speed, the unbalance of the winding tension distribution of the yarn is calculated, and the unbalance of the winding tension distribution is obtained. Based on the unbalanced winding tension distribution data, the yarn twist unevenness data is derived.

4. The yarn running control method according to claim 3, characterized in that, The abnormal set refers to the portion of the change amplitude in each time period that exceeds the tension reference threshold.

5. The yarn running control method according to claim 3, characterized in that, The calculation of yarn winding tension distribution imbalance includes: The concentricity center alignment offset is calculated based on the spindle belt tension and spindle speed. The discrete fluctuation of the tensile force of the spindle strip during yarn winding is estimated based on the alignment offset. The variance of the spatial orientation distribution of yarn tension during the winding process is calculated based on the discrete fluctuation of tensile force and the variance of abnormal tension changes. The unbalance of the winding tension distribution of yarn is calculated based on the variance of the spatial orientation distribution of tension, and the unbalanced winding tension distribution data is obtained.

6. The yarn running control method according to claim 2, characterized in that, Optimization of molding process conditions includes: Obtain the actual processed fineness of the yarn; The traveler running radius is adjusted based on the yarn twist unevenness data and the actual processed fineness. The intensity of centrifugal force imbalance is determined by the intensity of fluctuation; the height of the air ring is adjusted based on the intensity of centrifugal force imbalance, yarn twist unevenness data and actual processed fineness, and the adjustment range is controlled between 180mm-220mm, and the air ring height adjustment data is output. The forming process conditions are optimized based on the operating radius and air ring height adjustment data of the wire ring to obtain optimized process condition data.

7. The yarn running control method according to claim 2, characterized in that, Adaptive tension control adjustment of yarn includes: Set the initial tension control parameters, and control the initial tension between 0.06 and 0.08 CN / dtex; Based on yarn twist unevenness data and process condition optimization data, the yarn angle range was optimized and adjusted, including the yarn contact point angle range and the yarn tilt angle range. Based on the optimized process conditions and yarn angle range, the range of intervention force for tension control in the twisting zone is derived, and the range of intervention force for tension control in the twisting zone needs to be constrained to between 0.10-0.15 cN / dtex. Based on the tension intervention level data, the initial tension control parameters are adjusted for yarn adaptive tension control, and the tension control adjustment data is output.

8. The yarn running control method according to claim 1, characterized in that, Step S3 includes: Iterative simulation tests were conducted on the tension control adjustment data, and test data were collected. Perform boundary condition verification based on test data and output test condition verification data; Based on the test conditions, the data is verified and the tension control adjustment data is optimized. Then, the data is encoded and packaged to output a graded control coded data packet, which is sent to the control center of the twisting machine to execute yarn operation control.

9. A yarn running control system, characterized in that, For performing the yarn running control method as described in claim 1, the yarn running control system includes: The sample sampling module is used to extract yarn twisting process failure records from the control center of the twisting machine. The yarn twisting process failure records include tension fluctuation data obtained from the tension sensor. The module performs record sampling on the yarn twisting process failure records to obtain twisting failure sampling samples. The graded control module is used to deduce the yarn twist unevenness based on the failed twisting sample, and obtain the yarn twist unevenness data; based on the yarn twist unevenness data, it performs adaptive tension control adjustment of the yarn and outputs tension control adjustment data; The encoding module is used to optimize tension control adjustment data, and then encode and package the data to output graded control encoding data packets, which are then sent to the control center of the twisting machine to execute yarn running control.

Images