Stranded wire monitoring method and system of stranding machine

Abstract

The invention belongs to the technical field of cable manufacturing equipment monitoring, and provides a stranded wire monitoring method and system for a stranding machine, and the method comprises the steps: analyzing a sensor signal, judging whether a current pitch measurement signal has periodic jump or not, if yes, positioning the number of a single wire shielding a sensor and the corresponding shielding time in real time, and if not, determining that the current pitch measurement signal has periodic jump; the method comprises the following steps: establishing a normal reference of a single-wire jump value, determining collaborative characteristics of the jump value and a time parameter through actual and theoretical comparative analysis of the jump value and the time parameter, judging that a real pitch is abnormal, and converting a jump value offset into a pitch abnormal value based on a multiple linear regression model. The abnormal pitch critical value corresponding to the abnormity of the stranded wire is determined in combination with historical data, and finally the occurrence time of the abnormity of the stranded wire is predicted through the growth trend of the deviation of the single wire jump value, so that accurate monitoring and early warning from a single wire microscopic signal to a stranded wire macroscopic fault are realized, the accuracy of pitch abnormity judgment is improved, and invalid early warning is reduced.

Description

Technical Field

[0001] This invention belongs to the field of cable manufacturing equipment monitoring technology, specifically a method and system for monitoring stranded wires in a stranding machine. Background Technology

[0002] Stranding machines are core equipment in the production of industrial products such as wires and cables, steel wire ropes, optical cables, and steel strands. They use a rotating cage to twist multiple monofilaments at a preset helical angle to form strands. The pitch (the distance a monofilament travels in one revolution around the strand axis) is a key process parameter determining the stability, mechanical strength (such as tensile and torsional resistance), and service life of the stranded wire. Abnormal pitch (too large or too small) directly leads to: a loose stranded structure, uneven stress between monofilaments, easy localized wear or breakage, dimensional deviations in finished products that fail to meet assembly or usage standards, and in extreme cases, breakage due to excessive friction between monofilaments, causing production interruptions and batch scrap.

[0003] Existing methods for monitoring stranding machine pitch have the following main shortcomings: Traditionally, manual measurement or offline instrument sampling is used after the machine is stopped, which cannot reflect pitch changes during the production process in real time. By the time an anomaly is detected, a large number of defective products have already been generated. The detection frequency is low, making it difficult to capture instantaneous pitch fluctuations. It relies on manual operation, resulting in low efficiency and large errors. Some equipment attempts to collect signals online through sensors, but the core technical problems have not been solved: signal interference is difficult to eliminate; individual differences in single filaments can cause sensor signal jumps, which are easily misjudged as pitch anomalies, generating a large number of invalid warnings. Fluctuations in the winding speed will change the timing pattern of single filaments passing through the sensor. Existing methods do not dynamically adjust the time base, resulting in decreased monitoring accuracy. They can only determine whether there is an anomaly, but cannot quantify the specific value of the pitch deviation, let alone predict the time of failure based on the anomaly trend, making it difficult to support preventive maintenance. When multiple single filaments pass through the sensor alternately, existing technologies do not clearly distinguish the single filament numbers and cannot establish a correspondence between single filaments and signal characteristics, making it difficult to trace the source of anomalies.

[0004] Therefore, the present invention provides a method and system for monitoring stranded wires in a stranding machine. Summary of the Invention

[0005] In order to overcome the shortcomings of the prior art, at least one technical problem raised in the background art is solved.

[0006] The technical solution adopted by this invention to solve its technical problem is: a method for monitoring the stranded wires of a stranding machine, comprising: Step 1: Perform time interval stability analysis on the jump points in the current stranding machine sensor signal, determine the jump period, and match it with the theoretical period to determine whether the current pitch measurement signal has periodic jumps; Step 2: If this occurs, determine the filament number and blocking time of the blocking sensor based on the current stranding speed of the stranding machine and the arrangement pattern of the filaments. Step 3: Within the current stranding cycle, compare the actual and theoretical occlusion and occlusion intervals of the single filaments based on the filament number and occlusion time. Combine this with the comparison and analysis of the actual jump value of the single filament with the normal range of jump values ​​to determine the coordinated characteristics of the single filaments in the jump and time dimensions, and determine whether the periodic jump of the current pitch measurement signal is caused by an abnormality in the actual pitch. Step 4: If so, establish a mapping model of jump offset-pitch deviation based on historical twisting data, and calculate the true pitch anomaly value of the filament at the current moment by combining the jump value offset of the current filament. Step 5: Determine the critical value of pitch abnormality based on the abnormal pitch value corresponding to the abnormality of the stranded wire under the same working conditions, and predict the remaining time for the abnormality of the stranded wire by combining the mapping model of jump offset-pitch deviation and the actual pitch abnormal value.

[0007] Furthermore, the method for determining whether the current pitch measurement signal exhibits periodic jumps is as follows: Calculate the first-order difference sequence of the current sensor's measurement signal: ∆x i =|x i -x i-1 |(i≥2),∆x i x represents the signal change between two adjacent points. i This represents the value of the i-th measured signal; The x corresponding to the element in the first-order difference sequence that is greater than the transition threshold. i Mark the transition point and record the transition time; All transition points are integrated into a transition point sequence, and the standard deviation of the time interval between all adjacent transition points is calculated. If the standard deviation is less than the preset standard deviation, it is preliminarily judged that there is periodicity. Calculate the theoretical monofilament blocking period of the stranding machine: T 理论 =60 / (n·k), where n is the twisting speed and k is the number of single filaments. The average time interval between adjacent jump points is calculated and used as the actual jump period of the current pitch measurement signal. The deviation between the theoretical monofilament occlusion period and the actual jump period is calculated. If the deviation is less than the preset deviation, the current pitch measurement signal will show a periodic jump.

[0008] Furthermore, the method for determining the monofilament number of the occlusion sensor and the corresponding occlusion time is as follows: Obtain the current winding speed and the total number of filaments k involved in winding, and determine the ascending order of the filament numbers according to the direction of winding. The 360 ​​degrees of the winch rotation are divided into k equidistant intervals, each interval having an angle of 360° / k, and each interval corresponds to the shading period of a single filament. Calculate the current angle interval based on the rotation direction and the current angle θ, and determine the monofilament number based on the angle interval: Where k is the total number of monofilaments; The current time is t0, and the absolute angle output by the encoder in real time is θ(t0). The angle difference between the current angle and the starting point of the single-filament i-shaped occlusion interval is ∆θ=θ. start,i -θ(t0); The time difference corresponding to the angular difference is ∆t=(∆θ / 360°)·T 周期 If t is the time required to rotate by the angle ∆θ, then the occlusion start time of monofilament i is: t start,i =t0+∆t, ending time is t end,i =t start,i +T 遮挡,i .

[0009] Furthermore, the method for determining whether the periodic jumps in the current pitch measurement signal are caused by an abnormality in the actual pitch is as follows: For each monofilament involved in the twisting, suspected abnormal monofilaments are identified by comparing the actual jump value of the monofilament with the normal range of jump values. For each suspected abnormal monofilament; If the actual jump value of a single filament is greater than the upper limit of the normal jump value range, then the direction of the jump value offset of the single filament is marked as too large. If the actual jump value of a single filament is less than the lower limit of the normal jump value range, then the direction of the jump value offset of the single filament is marked as too small; If the jump values ​​of all suspected abnormal monofilaments are either too large or too small, then the jump values ​​are determined to be in the same direction. Calculate the deviation between the blocking time and the theoretical blocking time in the current twisting cycle, as well as the deviation between the blocking interval time of adjacent monofilaments and the theoretical interval time; If the deviation in duration of occlusion and the deviation in interval between adjacent monofilament occlusions are both greater than the preset deviation, and the deviations are in the same direction, either too long or too short, then the time offset direction is determined to be consistent. Suspected abnormal monofilaments with the same jump value offset direction and the same time offset direction were screened out, and the deviation of jump value and time was verified collaboratively. If the direction of the jump value offset is relatively large and the direction of the time offset is relatively long, or the direction of the jump value offset is relatively small and the direction of the time offset is relatively short, then the direction of the jump value offset and the direction of the time offset are coordinated, and the monofilament is marked as a coordinated monofilament. The proportion of cooperating monofilaments among all monofilaments is calculated to obtain the proportion of cooperating monofilaments. If the proportion of cooperating monofilaments is greater than the preset proportion, it is determined that the periodic jump in the current pitch measurement signal is caused by an abnormality in the actual pitch.

[0010] Furthermore, the method for identifying suspected abnormal transitions is as follows: Based on the 3σ principle, the upper and lower limits of the normal jump value of each single wire under the working conditions of stable operation of the stranding machine and no abnormal pitch are calculated to obtain the normal range of the jump value of the single wire. Before the single filament blocks the sensor, the sensor is in an unblocked gap period. During the gap period, the sensor signal is collected and the average value is calculated as the stable signal value before the single filament blocks the sensor. When collecting the signal peak value of the single-filament occlusion sensor, if it is a laser sensor, the signal peak value is the minimum value within the signal range; if it is an optical reflection sensor, the signal peak value is the maximum value within the signal range. Calculate the actual jump value ∆S of a single filament i =|S 峰，i -S 稳定 |, where S 峰，i S represents the peak signal value when a single filament i blocks the sensor. 稳定 The stable signal value before the sensor is blocked by a single filament; If the actual jump value of a single filament is ∆S i If the jump value of a single filament is outside the normal range, it is marked as a suspected abnormal jump.

[0011] Furthermore, the construction process of the mapping model for the jump offset-pitch deviation is as follows: Under known pitch deviation conditions, historical winding data are recorded synchronously, including: pitch deviation ∆H. 历史 =H 历史 -H0, where H0 is the standard pitch, H 历史 The actual pitch measured offline, and the offset ∆S of the jump value of each monofilament. 历史,i , that is, the deviation between the jump value of the i-th monofilament under this historical working condition and the normal jump average value; The multiple linear regression model is adopted, and the expression is ∆H=k1·∆S ’ 1+k2·∆S ’ 2+...+k k ·∆S ’ k +b, where k1~k k is the weighting coefficient for each monofilament, and b is a constant term; The historical twisting data was fitted using the least squares method to determine k1~k k And b, finally we get the mapping model of jump offset - pitch deviation.

[0012] Furthermore, the calculation method for the true pitch outlier is as follows: Obtain the jump value offset of all monofilaments, and calculate the contribution of each monofilament to the pitch deviation using the formula: Contribution i =|k i |×|∆S i |, where k i It is the weight coefficient of the i-th monofilament in the multiple linear regression model, |∆S i | is the absolute value of the jump value offset of the i-th monofilament in the current monitoring; Single filaments with a contribution greater than the preset contribution are marked as major contributing single filaments; Substituting the jump value offset of the main contributing monofilament into the multiple linear regression model, the current pitch deviation value is obtained; The current pitch deviation value is added to the standard pitch to obtain the current true pitch anomaly value.

[0013] Furthermore, the method for determining the pitch anomaly critical value is as follows: Obtain the abnormal pitch values ​​of all stranded wire abnormalities during historical stranding processes to form an abnormal sample set; Calculate the coefficient of variation of the abnormal sample set. If the coefficient of variation is less than the preset coefficient of variation, calculate the mean of the abnormal sample set as the critical value of the pitch abnormality corresponding to the twisted wire abnormality. Otherwise, select the minimum value in the abnormal sample set as the critical value of the pitch abnormality corresponding to the twisted wire abnormality.

[0014] Furthermore, the prediction process for the remaining time of the stranded wire abnormality is as follows: Calculate the difference ∆H between the current true pitch anomaly value and the pitch anomaly threshold value; Extract the jump value offset ∆S of each monofilament from historical twisting data. 历史,i The jump value offset sequence is obtained: {∆S ’ i (t1), ∆S ’ i (t2), ...,∆S ’ i (t n )}, where t1 <t2<...<t n , is the timestamp, ∆S ’ i (t n ) represents the single filament i at timestamp t n The offset of the corresponding jump value; By linearly fitting the change in the jump value offset sequence over time, a fitting model is obtained: ,in, c represents the growth rate of the offset of the jump value of the i-th monofilament. i For constant terms; The growth rate of the jump value offset of the i-th monofilament is obtained by solving the least squares method. ; The growth rate of the pitch outlier is the weighted sum of the growth rates of the jump values ​​of each monofilament, with the weights being the coefficients k in the jump offset-pitch deviation mapping model. i The specific weighting formula is as follows: ,in, v represents the rate of increase of the jump value offset of the i-th monofilament. H This represents the growth rate of the pitch anomaly. The remaining time for the current true pitch anomaly to reach the pitch anomaly threshold is , where ∆H is the difference between the current true pitch anomaly value and the anomaly threshold value.

[0015] A strand monitoring system for a stranding machine, comprising the following modules: Periodic jump identification module: Performs time interval stability analysis on the jump points in the current stranding machine sensor signal, determines the jump period, and performs matching analysis with the theoretical period to determine whether the current pitch measurement signal has a periodic jump; Single filament obstruction determination module: If obstruction occurs, the single filament number and obstruction time of the obstruction sensor are determined based on the current stranding speed of the stranding machine and the single filament arrangement pattern. Actual pitch anomaly judgment module: Within the current stranding cycle, the module compares the actual and theoretical occlusion and occlusion interval of the single filament based on the single filament number and occlusion time, and combines the comparison and analysis of the actual jump value of the single filament with the normal range of jump value to determine the coordinated characteristics of the single filament in the jump and time dimensions, and judges whether the periodic jump of the current pitch measurement signal is caused by an actual pitch anomaly. Pitch anomaly calculation module: If so, establish a mapping model of jump offset - pitch deviation based on historical twisting data, and calculate the real pitch anomaly of the filament at the current moment by combining the jump value offset of the current filament. Stranded wire anomaly early warning module: Based on the abnormal pitch value corresponding to the stranded wire anomaly under the same working conditions, the critical value of pitch anomaly is determined, and combined with the mapping model of jump offset-pitch deviation and the actual pitch anomaly value, the remaining time of stranded wire anomaly is predicted.

[0016] The beneficial effects of this invention are as follows: It accurately identifies periodic jumps through time-domain standard deviation, periodic matching, and frequency-domain peak value verification, effectively eliminating non-periodic interference and providing a reliable signal benchmark for subsequent analysis. It achieves real-time monitoring of strand pitch, enabling timely detection of pitch anomalies and avoiding the generation of numerous defective products due to delayed detection. By distinguishing between individual filament anomalies and true pitch anomalies, it eliminates the interference of filament defects on pitch judgment, improving the accuracy of pitch anomaly judgment and reducing invalid warnings. It establishes a mapping model between filament jump value offset and pitch deviation, transforming abstract sensor signals into physically interpretable pitch anomaly values. This achieves a quantitative correlation between signal characteristics and process parameter anomalies, enabling the prediction of the time when accumulated pitch anomalies eventually lead to strand anomalies, providing a decision-making basis for equipment maintenance, facilitating planned maintenance, reducing production interruptions caused by sudden failures, and exhibiting good adaptability to speed fluctuations. By updating speed parameters in real time, it ensures synchronization between angle interval division and actual rotation cycle, improving monitoring accuracy under different operating conditions. Attached Figure Description

[0017] The invention will now be further described with reference to the accompanying drawings.

[0018] Figure 1 This is a flowchart of the steps of a strand monitoring method for a stranding machine according to Embodiment 1 of the present invention; Figure 2 This is a logic judgment diagram of a strand monitoring method for a stranding machine according to Embodiment 1 of the present invention; Figure 3 This is a flowchart of a strand monitoring system for a stranding machine according to Embodiment 2 of the present invention. Detailed Implementation

[0019] To make the technical means, creative features, objectives and effects of this invention easier to understand, the invention will be further described below in conjunction with specific embodiments.

[0020] Example 1: Please refer to Figure 1 As shown in the embodiment of the present invention, a method for monitoring the stranded wires of a stranding machine includes the following steps: Step 1: Perform time interval stability analysis on the jump points in the current stranding machine sensor signal, determine the jump period, and match it with the theoretical period to determine whether the current pitch measurement signal has periodic jumps; In step one, the process of determining whether the current pitch measurement signal exhibits periodic jumps includes: Sensor signals are continuously acquired at a sampling rate higher than the monofilament occlusion frequency, and the first-order difference sequence of the sensor signals is calculated: ∆x i =|x i -x i-1|(i≥2),∆x i x represents the signal change between two adjacent points. i This represents the value of the i-th measured signal; Each element in the first-order difference sequence is compared with the transition threshold. If ∆x i If the value is greater than the transition threshold, x will be... i Mark the point as a transition point and record the transition time and transition magnitude |x. i -x i-1 |; Integrate all transition points to generate a transition point sequence P=[p1,p2,...,p k ], where k is the number of transition points, and the time interval between adjacent transition points is calculated, ∆t1,∆t2,...,∆t i ; Calculate the standard deviation of the time interval between adjacent jump points and compare it with the preset standard deviation. If the standard deviation is less than the preset standard deviation, it is preliminarily judged that there is periodicity. Calculate the theoretical monofilament blocking period of the stranding machine: T 理论 =60 / (n·k), calculate the average time interval between adjacent transition points, and use it as the actual transition period T of the current pitch measurement signal. 实际 ; Calculate the deviation between the theoretical monofilament occlusion period and the actual jump period, and compare it with the preset deviation. If the deviation is less than the preset deviation, it means that the signal jump period matches the theoretical period of the twisting motion, and the current pitch measurement signal has a periodic jump. The purpose of determining whether the current pitch measurement signal exhibits periodic jumps is: Function 1: Verify periodicity from multiple dimensions, effectively eliminate the influence of non-periodic interference such as sensor noise and instantaneous vibration, and avoid misjudging random fluctuations as normal periodic jumps; Function 2: Accurately identify whether the pitch measurement signal has periodic jumps due to alternating blockage of single filaments, laying the foundation for distinguishing between normal periodic jumps and abnormal signal fluctuations. Step 2: If this occurs, determine the filament number and blocking time of the blocking sensor based on the current stranding speed of the stranding machine and the arrangement pattern of the filaments. In step two, the process of determining the monofilament number of the occlusion sensor and the corresponding occlusion time includes: The encoder collects the winding speed n in real time, in revolutions per minute. Based on the design parameters of the stranding machine, the total number k of the monofilaments involved in stranding is determined and recorded. Manually rotate the winch to the position where the No. 1 monofilament is directly opposite the sensor monitoring point, and set this position as the zero angle point, i.e., 0 degrees, through the encoder software; Four magnets are evenly installed on the end face of the winch, and four Hall sensors are installed at corresponding positions on the fixed bracket. When the winch rotates, the sensors are triggered in sequence, and the rotation direction is determined by the triggering sequence. For example, if the four Hall sensors A, B, C, and D rotate clockwise, they will be triggered in sequence: A→B→C→D, and counterclockwise, they will be triggered in sequence: D→C→B→A. The ascending order of the monofilament numbers is determined by the direction of rotation. For example, when rotating clockwise, the monofilaments are numbered 1→2→3→4, and when rotating counterclockwise, they are numbered 1→4→3→2. The time it takes for the winch to rotate one revolution is T. 周期 =60 / n, where n is the winding speed, and the time each single filament blocks the sensor is equal to the time corresponding to its angle interval, i.e.: T 遮挡,i =T 周期 / k, where T 遮挡,i The time when the i-th monofilament blocks the sensor; The 360 ​​degrees of the winch rotation are divided into k equidistant intervals, each interval having an angle of 360° / k. Each interval corresponds to the blocking period of a single filament. For example, 0°~360° / k corresponds to filament No. 1, 360° / k~2×360° / k corresponds to filament No. 2, and so on. The encoder outputs the absolute angle θ (in degrees) of the coil rotation in real time. Since the encoder resolution is the number of pulses per revolution, the pulse count needs to be converted into an angle value. ; Calculate the current angle interval based on the rotation direction and the current angle θ, and determine the monofilament number based on the angle interval: For example: if the winch rotates clockwise and θ = 120°, for k = 7, then the interval index is... This corresponds to monofilament number 3 (because the index starts from 0); The current time is t0, and the absolute angle output by the encoder in real time is θ(t0). The occlusion start time t of the current monofilament i is calculated based on the angle. start,i and end time t end,i The angle difference between the current angle and the starting point of the occlusion interval of the single filament i is ∆θ=θ start,i -θ(t0); The time difference corresponding to the angular difference is ∆t=(∆θ / 360°)·T 周期 That is, the time required to rotate by the angle ∆θ. The occlusion start time of monofilament i is: t start,i =t0+∆t, ending time is t end,i =t start,i +T 遮挡,i ; When the speed n of the winch fluctuates due to acceleration / deceleration, n is updated in real time by the encoder, and the theoretical time interval between the single filament passing through the monitoring point is recalculated to ensure that the angle interval division is always synchronized with the actual rotation cycle. The purpose of determining the monofilament number of the occlusion sensor is: Function 1: By associating the filament number with time, it is possible to accurately locate when a filament blocks the sensor, providing a time reference for subsequent acquisition of stable signal values ​​before filament blockage and capture of signal peak values ​​at the time of blockage, thus avoiding signal acquisition time deviation caused by rotation speed fluctuations. Function 2: It solves the problem of identity confusion caused by alternating occlusion of multiple monofilaments, provides a clear monofilament benchmark for subsequent jump value analysis, avoids signal interpretation errors caused by misjudgment of monofilament identity, and is the basis for realizing pitch anomaly monitoring; Step 3: Within the current stranding cycle, compare the actual and theoretical occlusion and occlusion intervals of the single filaments based on the filament number and occlusion time. Combine this with the comparison and analysis of the actual jump value of the single filament with the normal range of jump values ​​to determine the coordinated characteristics of the single filaments in the jump and time dimensions, and determine whether the periodic jump of the current pitch measurement signal is caused by an abnormality in the actual pitch. Please see Figure 2 As shown, in step three, the process of determining whether the periodic jumps in the current pitch measurement signal are caused by an abnormality in the actual pitch is as follows: Under the condition that the stranding machine is running stably and the pitch is normal, the following are recorded simultaneously: the single wire number of the sensor that is blocked, the jump value of the pitch measurement signal when the single wire blocks the sensor, that is, the difference between the stable value of the signal before blocking and the peak value of the signal when blocking. The recorded data are grouped by filament number, and the upper and lower limits of the normal jump value for each filament are calculated based on the 3σ principle to obtain the normal range of the jump value for each filament. A mapping table of filament number and normal jump range is established: |filament number|normal jump range[L i U i ]|; Before the filament i is blocked, the sensor is in the unblocked gap period. At this time, the sensor is not blocked and the signal is stable without jumps. The sensor signal S(t) is collected during the gap period and the average value is calculated as the stable signal value before the filament i is blocked. When a single filament i blocks the sensor, due to the change in the profile of the single filament helical surface (passing through the closest / farthest point from the sensor), the signal S(t) will have a peak. The signal peak value when the single filament blocks the sensor is collected. If it is a laser sensor, the signal is the smallest when the single filament is closest, so the signal peak value is the minimum value in the signal interval. If it is an optical reflection sensor, the signal is the largest when the single filament reflection is the strongest, so the signal peak value is the maximum value in the signal interval. Calculate the actual jump value ∆S of a single filament 当前,i =|S 峰，i-S 稳定 |, where S 峰，i S represents the peak signal value when a single filament i is blocked. 稳定 The stable signal value before monofilament i is blocked; For each filament i, the theoretical occlusion time of the filament and the theoretical time interval between adjacent filaments are both T. 理论, i=60 / (n·k), because when the winch rotates at a constant speed, the monofilaments are evenly distributed, and the time interval between adjacent blocking is the same as the blocking time of the monofilament itself. For the currently monitored monofilament, calculate the actual occlusion time of the monofilament and the actual time interval between adjacent monofilaments within the current twisting cycle; For each filament of the occlusion sensor, the actual jump value ∆S of the filament is... 当前,i Compare with the normal range of jump values; If the actual jump value of a single filament is ∆S 当前,i Within the normal range of jump values ​​[L] i U i Within [the range], it is determined to be a normal shading jump, caused by the inherent characteristics of the monofilament, and not a pitch abnormality; Conversely, if the change is not observed, it is considered a suspected abnormal transition, and the monofilament is marked as a suspected abnormal monofilament. It should be noted that suspected anomalous jumps may be caused by two reasons: Individual problems with a single filament: such as a sudden change in the diameter of the single filament (wear, joint) or a surface protrusion, which may cause abnormal signal jumps (unrelated to pitch). Abnormal actual pitch: The stranding pitch itself deviates from the rated value, causing a systematic shift in the blocking transition of all monofilaments (because pitch changes will change the overall shape of the monofilament helical surface, affecting the blocking signal of all monofilaments). The jump values ​​of all single filaments within the same twisting cycle are statistically analyzed and compared with their corresponding normal ranges. If only the current filament i is suspected of having an abnormal jump, while the actual jump values ​​of other filaments are all within the normal range, then it is an individual filament abnormality. Conversely, for each suspected abnormal filament, the offset direction is defined based on the relationship between the actual jump value of the filament and the normal range of jump values: If the actual jump value of a single filament is greater than the upper limit of the normal jump value range, then the direction of the jump value offset of the single filament is marked as too large. If the actual jump value of a single filament is less than the lower limit of the normal range, then the direction of the jump value offset of the single filament is marked as too small; Iterate through all suspected abnormal monofilaments, record the offset direction of the jump value corresponding to each suspected abnormal monofilament, and form a set of offset directions; If the jump values ​​of all suspected abnormal monofilaments in the set are either too large or too small, then the jump values ​​are determined to be in the same direction. For each suspected abnormal filament, calculate the deviation between the occlusion time and the theoretical occlusion time in the current twisting cycle, as well as the deviation between the occlusion interval time of adjacent filaments and the theoretical interval time, and compare them with the preset deviation. If the deviation of the occlusion duration is greater than the preset deviation, and the deviation of the occlusion interval time of adjacent filaments is greater than the preset deviation, and the deviation direction is the same, that is, either too long or too short, then the time offset direction is determined to be consistent. Suspected abnormal monofilaments with consistent jump value and time offset directions were identified, and the deviations in jump value and time were jointly verified. If the direction of the jump value offset is relatively large and the direction of the time offset is relatively long, or if the direction of the jump value offset is relatively small and the direction of the time offset is relatively short, then the offset directions of the jump value and the time parameter are coordinated, and the monofilament is marked as a coordinated monofilament. The proportion of cooperating monofilaments in all monofilaments is statistically analyzed to obtain the proportion of cooperating monofilaments, and then compared with the preset proportion. If the proportion of cooperating monofilaments is greater than the preset proportion, it is determined that the actual pitch is abnormal. It should be noted that the judgment logic of true pitch anomaly is as follows: true pitch anomaly will affect both the amplitude and temporal pattern of the jump value of a single filament (occlusion time, adjacent interval), while individual filament anomaly only affects the amplitude of the jump value and does not change the temporal pattern. By comparing the pitch measurement signal jump value caused by the current monofilament occlusion with the pitch measurement signal jump value normally generated by the monofilament, the function of determining whether the periodic jump in the current pitch measurement signal is caused by an abnormality in the actual pitch is as follows: Function 1: By comparing the current monofilament jump value with the historical normal reference range, it distinguishes between normal shading jump, individual monofilament anomaly and true pitch anomaly, and eliminates the interference of monofilament defects (such as abrupt diameter change or surface protrusion) on pitch judgment. Function 2: Improve the accuracy of pitch anomaly judgment - only when the jump values ​​of multiple single filaments are systematically deviated (the proportion exceeds the preset value and the direction is consistent) is it judged as a real anomaly, avoiding misjudging individual problems of single filaments as pitch anomalies and reducing invalid warnings; Step 4: If so, establish a mapping model of jump offset-pitch deviation based on historical twisting data, and calculate the true pitch anomaly value of the filament at the current moment by combining the jump value offset of the current filament. In step four, the calculation process for the current true pitch anomaly value includes: Under known pitch deviation conditions, historical winding data are recorded synchronously, including: pitch deviation ∆H. 历史 =H 历史 -H0, where H0 is the standard pitch, H 历史 The actual pitch measured offline, and the offset ∆S of the jump value of each monofilament. 历史,i , that is, the deviation between the jump value of the i-th monofilament under this historical working condition and the normal jump average value; The multiple linear regression model is adopted, and the expression is ∆H=k1·∆S ’ 1+k2·∆S ’ 2+...+k k ·∆S ’ k +b, where k1~k k is the weighting coefficient for each filament, reflecting the sensitivity of the filament's jump offset to the pitch deviation, and b is a constant term; Historical data were fitted using the least squares method to determine k1~k k And b, finally we get the mapping model of jump offset - pitch deviation; For current monitoring data identified as true pitch anomalies, extract the jump value offset of all filaments and calculate the contribution of each filament to the pitch deviation. The formula is: Contribution. i =|k i |×|∆S i |, where k i It is the weight coefficient of the i-th monofilament in the historical multiple linear regression model, |∆S i | is the absolute value of the jump value offset of the i-th monofilament in the current monitoring; Understandably, the physical meaning of contribution is: under the current working conditions, the combined influence of the monofilament's sensitivity to pitch changes, plus its actual path deviation, on the overall pitch anomaly, quantifies the relationship between the monofilament and the pitch anomaly from two dimensions. Wherein, the weighting coefficient represents the inherent correlation strength between the jump value shift of a single filament and the pitch deviation, if |k i A larger value indicates that the jump value deviation of a single filament is highly sensitive to pitch deviation. Even if the jump value deviation of the single filament is small, it may correspond to a large pitch anomaly. The absolute value of the jump value deviation represents the magnitude of the jump value deviation of the single filament in the current monitoring. The larger the deviation, the more significant the actual impact on the pitch deviation. The contribution of each monofilament is compared with the preset contribution, and the monofilaments with a contribution greater than the preset contribution are marked as the main contributing monofilaments. Substituting the jump value offset of the main contributing monofilament into the multiple linear regression model, we obtain the current pitch deviation value, ∆H. 预测 =k1·∆S ’ 1+k2·∆S ’ 2+...+k k ·∆S ’ k +b; The current pitch deviation value is obtained by adding the pitch deviation value to the standard pitch, where the standard pitch is obtained based on the stranding machine design parameters; The purpose of calculating the current true pitch outlier is: Function 1: Based on historical data, a jump value offset-pitch deviation mapping model is established to quantify the current monofilament jump value offset into pitch anomalies, and key monofilaments are screened by contribution to improve calculation accuracy; Function 2: The abstract signal from the sensor is transformed into a pitch anomaly value that can be physically interpreted, realizing the quantitative correlation between signal characteristics and process parameter anomalies, and providing specific and traceable anomaly indicators for subsequent early warning; Step 5: Determine the critical value of pitch abnormality based on the abnormal pitch value corresponding to the abnormality of the stranded wire under the same working conditions, and predict the remaining time for the abnormality of the stranded wire by combining the mapping model of jump offset-pitch deviation and the actual pitch abnormal value.

[0021] In step five, the stranding abnormality refers to the continuous occurrence of abnormal actual pitch in the stranding machine. Due to the continuous deviation of the single wire spiral path (such as excessively large / small pitch leading to abnormal spiral angle), the relative position of the single wire and the sensor, and the force relationship between the single wires will gradually deteriorate, which may eventually lead to problems such as excessive wear or breakage of single wires, loose stranding or untwisted strands, severe degradation of mechanical properties, and abnormal wear of equipment components. The process for determining the critical value of the pitch anomaly includes: Obtain the abnormal pitch value corresponding to the abnormal stranded wire caused by the abnormal pitch value under the same historical working conditions, and record the specific time when the abnormality occurred; The pitch values ​​corresponding to all abnormal moments of stranded wires were selected from historical data to form an abnormal sample set; Calculate the coefficient of variation of the abnormal sample set and compare it with the preset coefficient of variation. If the coefficient of variation is less than the preset coefficient of variation, the pitch outlier is stable; otherwise, it is unstable. If the pitch anomaly is stable, the mean of the anomaly sample set is calculated as the pitch anomaly threshold value corresponding to the twisted wire anomaly. Conversely, if it is unstable, the minimum value in the anomaly sample set is selected as the pitch anomaly threshold value corresponding to the twisted wire anomaly. Understandably, the logic for determining the critical value of pitch anomalies based on their stability is as follows: if the pitch anomaly is stable, it means that the pitch anomaly fluctuated little during historical faults, and using the mean as the critical value is more representative and can reflect the common threshold of most fault scenarios. Conversely, if it is unstable, it means that the pitch anomaly fluctuated greatly during historical faults, and using the minimum value as the critical value is more conservative and can avoid the critical value being too high due to data fluctuations, thus providing early warning. In step five, the prediction process for the remaining time of the stranded wire abnormality includes: Obtain the real-time jump value offset of each monitored filament, as well as the time-series data of the jump value offset within the historical period {∆S}. ’ i (t)}; Calculate the difference ∆H between the current true pitch anomaly value and the pitch anomaly threshold value; From historical periodic monitoring data, extract the jump value offset sequence of the i-th monofilament: {∆S ’ i (t1), ∆S ’ i (t2), ...,∆S ’ i (t n )}, where t1 <t2<...<t n , is the timestamp, ∆S ’ i (t n ) represents the single filament i at timestamp t n The offset of the corresponding jump value; By linearly fitting the change of the jump value offset sequence over time, the growth rate of the jump value offset of the i-th filament is obtained. The unit is the jump value unit / s, specifically: Let the time variable be t. Starting from t1, convert it to a relative time t' = t - t1, then the fitted model is: ,in, c represents the growth rate of the offset of the jump value of the i-th monofilament. i For constant terms; Solve using the least squares method ,Right now: ; According to the jump offset-pitch deviation mapping model, the growth rate of the pitch anomaly is the weighted sum of the growth rates of the jump offsets of each monofilament, with the weights being the coefficients k in the mapping model. i , ,in, The rate of increase of the jump value offset of the i-th monofilament; The remaining time for the current true pitch anomaly to reach the pitch anomaly threshold is Where ∆H is the difference between the current true pitch anomaly value and the anomaly threshold value, v H This represents the growth rate of the pitch anomaly. It should be noted that the deviation of the jump value of a single filament is a microscopic manifestation of the pitch anomaly. By aggregating the growth trend of the jump value of a single filament, the overall growth rate of the pitch anomaly can be obtained. Then, combined with the current difference from the critical value, the remaining time of the fault can be estimated by extrapolating the linear trend, so as to realize the time early warning from microscopic indicators to macroscopic faults. The purpose of predicting the time when the stranded wire will eventually experience an anomaly is: Function 1: Determine the critical value of pitch anomaly through historical fault data (distinguishing between stable and unstable scenarios), and then aggregate the overall growth rate of pitch anomaly based on the growth trend of single filament jump value offset, and finally estimate the remaining time to reach the critical value. Function 2: To achieve a closed loop from anomaly identification to fault early warning, to clarify the danger threshold of pitch anomalies, and to inform in advance of the possible time of faults by combining trend prediction, so as to provide a basis for decision-making for equipment maintenance (such as downtime adjustment and parameter correction) and reduce production interruption and scrap loss caused by sudden failures.

[0022] The technical solution and advantages of this application embodiment are as follows: Time interval stability analysis is performed on the transition points in the current stranding machine sensor signal to determine the transition period, and a matching analysis is performed with the theoretical period to determine whether the current pitch measurement signal exhibits periodic transitions. If so, based on the current stranding speed of the stranding machine and the single-wire arrangement pattern, the single-wire number and blocking time of the blocking sensor are determined. Within the current stranding period, the actual and theoretical blocking and blocking intervals of the single wire are compared based on the single-wire number and blocking time, and the actual transition value of the single wire is combined with the transition value... By comparing and analyzing within the normal range, the coordinated characteristics of the single filament in the jump and time dimensions are determined. It is then determined whether the periodic jump in the current pitch measurement signal is caused by an abnormality in the actual pitch. If so, a mapping model of jump offset-pitch deviation is established based on historical stranding data. The actual pitch abnormality value of the single filament at the current moment is calculated by combining the jump value offset of the current single filament. Based on the abnormal pitch value corresponding to the abnormality of the strand under the same working conditions, the critical value of pitch abnormality is determined. Finally, by combining the mapping model of jump offset-pitch deviation and the actual pitch abnormality value, the remaining time for the strand to become abnormal is predicted. This application analyzes sensor signals to determine whether periodic jumps occur in the current pitch measurement signal. If so, it locates the filament number and corresponding occlusion time of the obstructing sensor in real time, establishes a normal benchmark for the filament jump value, and determines the synergistic characteristics of the jump value and time parameter by comparing and analyzing the actual jump value and time parameter. Based on a multiple linear regression model, the jump value offset is converted into an abnormal pitch value, and combined with historical data, the critical value of the abnormal pitch corresponding to the strand anomaly is determined. Finally, the occurrence time of the strand anomaly is predicted by the growth trend of the filament jump value offset. This achieves accurate monitoring and early warning from microscopic signals of filaments to macroscopic faults of strands, improves the accuracy of pitch anomaly judgment, and reduces invalid warnings.

[0023] Example 2: Please refer to Figure 3 As shown in the embodiment of the present invention, a strand monitoring system for a stranding machine includes the following modules: Periodic jump identification module: Performs time interval stability analysis on the jump points in the current stranding machine sensor signal, determines the jump period, and performs matching analysis with the theoretical period to determine whether the current pitch measurement signal has a periodic jump; The process of determining whether the current pitch measurement signal exhibits periodic jumps includes: Sensor signals are continuously acquired at a sampling rate higher than the monofilament occlusion frequency, and the first-order difference sequence of the sensor signals is calculated: ∆x i =|x i -x i-1 |(i≥2),∆x i x represents the signal change between two adjacent points. i This represents the value of the i-th measured signal; Each element in the first-order difference sequence is compared with the transition threshold. If ∆x i If the value is greater than the transition threshold, x will be... i Mark the point as a transition point and record the transition time and transition magnitude |x. i -x i-1 |; Integrate all transition points to generate a transition point sequence P=[p1,p2,...,p k ], where k is the number of transition points, and the time interval between adjacent transition points is calculated, ∆t1,∆t2,...,∆t i ; Calculate the standard deviation of the time interval between adjacent jump points and compare it with the preset standard deviation. If the standard deviation is less than the preset standard deviation, it is preliminarily judged that there is periodicity. Calculate the theoretical monofilament blocking period of the stranding machine: T 理论 =60 / (n·k), calculate the average time interval between adjacent transition points, and use it as the actual transition period T of the current pitch measurement signal. 实际 ; Calculate the deviation between the theoretical monofilament occlusion period and the actual jump period, and compare it with the preset deviation. If the deviation is less than the preset deviation, it means that the signal jump period matches the theoretical period of the twisting motion, and the current pitch measurement signal has a periodic jump. Single filament obstruction determination module: If obstruction occurs, the single filament number and obstruction time of the obstruction sensor are determined based on the current stranding speed of the stranding machine and the single filament arrangement pattern. The process of determining the monofilament number of the occlusion sensor and the corresponding occlusion time includes: The encoder collects the winding speed n in real time, in revolutions per minute. Based on the design parameters of the stranding machine, the total number k of the monofilaments involved in stranding is determined and recorded. Manually rotate the winch to the position where the No. 1 monofilament is directly opposite the sensor monitoring point, and set this position as the zero angle point, i.e., 0 degrees, through the encoder software; Four magnets are evenly installed on the end face of the winch, and four Hall sensors are installed at corresponding positions on the fixed bracket. When the winch rotates, the sensors are triggered in sequence, and the rotation direction is determined by the triggering sequence. Determine the ascending order of the monofilament numbers by considering the direction of rotation; The time it takes for the winch to rotate one revolution is T. 周期 =60 / n, where n is the winding speed, and the time each single filament blocks the sensor is equal to the time corresponding to its angle interval, i.e.: T 遮挡,i =T 周期 / k, where T 遮挡,i The time when the i-th monofilament blocks the sensor; The 360 ​​degrees of the winch rotation are divided into k equidistant intervals, each interval having an angle of 360° / k, and each interval corresponds to the shading period of a single filament. The encoder outputs the absolute angle θ (in degrees) of the coil rotation in real time. Since the encoder resolution is the number of pulses per revolution, the pulse count needs to be converted into an angle value. ; Calculate the current angle interval based on the rotation direction and the current angle θ, and determine the monofilament number based on the angle interval: ; The current time is t0, and the absolute angle output by the encoder in real time is θ(t0). The occlusion start time t of the current monofilament i is calculated based on the angle. start,i and end time t end,i The angle difference between the current angle and the starting point of the occlusion interval of the single filament i is ∆θ=θ start,i -θ(t0); The time difference corresponding to the angular difference is ∆t=(∆θ / 360°)·T 周期 That is, the time required to rotate by the angle ∆θ. The occlusion start time of monofilament i is: t start,i =t0+∆t, ending time is t end,i =t start,i +T 遮挡,i ; When the speed n of the winch fluctuates due to acceleration / deceleration, n is updated in real time by the encoder, and the theoretical time interval between the single filament passing through the monitoring point is recalculated to ensure that the angle interval division is always synchronized with the actual rotation cycle. Actual pitch anomaly judgment module: Within the current stranding cycle, the module compares the actual and theoretical occlusion and occlusion interval of the single filament based on the single filament number and occlusion time, and combines the comparison and analysis of the actual jump value of the single filament with the normal range of jump value to determine the coordinated characteristics of the single filament in the jump and time dimensions, and judges whether the periodic jump of the current pitch measurement signal is caused by an actual pitch anomaly. Please see Figure 2As shown, the process of determining whether the periodic jumps in the current pitch measurement signal are caused by an abnormality in the actual pitch is as follows: Under the condition that the stranding machine is running stably and the pitch is normal, the following are recorded simultaneously: the single wire number of the sensor that is blocked, the jump value of the pitch measurement signal when the single wire blocks the sensor, that is, the difference between the stable value of the signal before blocking and the peak value of the signal when blocking. The recorded data are grouped by filament number, and the upper and lower limits of the normal jump value for each filament are calculated based on the 3σ principle to obtain the normal range of the jump value for each filament. A mapping table of filament number and normal jump range is established: |filament number|normal jump range[L i U i ]|; Before the filament i is blocked, the sensor is in the unblocked gap period. At this time, the sensor is not blocked and the signal is stable without jumps. The sensor signal S(t) is collected during the gap period and the average value is calculated as the stable signal value before the filament i is blocked. When a single filament i blocks the sensor, due to the change in the profile of the single filament helical surface (passing through the closest / farthest point from the sensor), the signal S(t) will have a peak. The signal peak value when the single filament blocks the sensor is collected. If it is a laser sensor, the signal is the smallest when the single filament is closest, so the signal peak value is the minimum value in the signal interval. If it is an optical reflection sensor, the signal is the largest when the single filament reflection is the strongest, so the signal peak value is the maximum value in the signal interval. Calculate the actual jump value ∆S of a single filament 当前,i =|S 峰，i -S 稳定 |, where S 峰，i S represents the peak signal value when a single filament i is blocked. 稳定 The stable signal value before monofilament i is blocked; For each filament i, the theoretical occlusion time of the filament and the theoretical time interval between adjacent filaments are both T. 理论, i=60 / (n·k), because when the winch rotates at a constant speed, the monofilaments are evenly distributed, and the time interval between adjacent blocking is the same as the blocking time of the monofilament itself. For the currently monitored monofilament, calculate the actual occlusion time of the monofilament and the actual time interval between adjacent monofilaments within the current twisting cycle; For each filament of the occlusion sensor, the actual jump value ∆S of the filament is... 当前,i Compare with the normal range of jump values; If the actual jump value of a single filament is ∆S 当前,i Within the normal range of jump values ​​[L] i U i Within [the range], it is determined to be a normal shading jump, caused by the inherent characteristics of the monofilament, and not a pitch abnormality; Conversely, if the change is not observed, it is considered a suspected abnormal transition, and the monofilament is marked as a suspected abnormal monofilament. The jump values ​​of all single filaments within the same twisting cycle are statistically analyzed and compared with their corresponding normal ranges. If only the current filament i is suspected of having an abnormal jump, while the actual jump values ​​of other filaments are all within the normal range, then it is an individual filament abnormality. Conversely, for each suspected abnormal monofilament, the offset direction is defined based on the relationship between the current jump value and the normal range of jump values: If the actual jump value of a single filament is greater than the upper limit of the normal jump value range, then the direction of the jump value offset of the single filament is marked as too large. If the actual jump value of a single filament is less than the lower limit of the normal range, then the direction of the jump value offset of the single filament is marked as too small; Iterate through all suspected abnormal monofilaments, record the offset direction of the jump value corresponding to each suspected abnormal monofilament, and form a set of offset directions; If the jump values ​​of all suspected abnormal monofilaments in the set are either too large or too small, then the jump values ​​are determined to be in the same direction. For each suspected abnormal filament, calculate the deviation between the occlusion time and the theoretical occlusion time in the current twisting cycle, as well as the deviation between the occlusion interval time of adjacent filaments and the theoretical interval time, and compare them with the preset deviation. If the occlusion time deviation is greater than the preset deviation, and the occlusion interval time deviation of adjacent filaments is greater than the preset deviation, and the deviation direction is the same, either too long or too short, then the time offset direction is determined to be consistent. Suspected abnormal monofilaments with consistent jump value and time offset directions were identified, and the deviations in jump value and time were jointly verified. If the direction of the jump value offset is relatively large and the direction of the time offset is relatively long, or if the direction of the jump value offset is relatively small and the direction of the time offset is relatively short, then the offset directions of the jump value and the time parameter are coordinated, and the monofilament is marked as a coordinated monofilament. The proportion of cooperating monofilaments in all monofilaments is statistically analyzed to obtain the proportion of cooperating monofilaments, and then compared with the preset proportion. If the proportion of cooperating monofilaments is greater than the preset proportion, it is determined that the actual pitch is abnormal. Pitch anomaly calculation module: If so, establish a mapping model of jump offset - pitch deviation based on historical twisting data, and calculate the real pitch anomaly of the filament at the current moment by combining the jump value offset of the current filament. The calculation process for the current true pitch anomaly value includes: Under known pitch deviation conditions, historical winding data are recorded synchronously, including: pitch deviation ∆H. 历史 =H 历史 -H0, where H0 is the standard pitch, H 历史 The actual pitch measured offline, and the offset ∆S of the jump value of each monofilament. 历史,i , that is, the deviation between the jump value of the i-th monofilament under this historical working condition and the normal jump average value; The multiple linear regression model is adopted, and the expression is ∆H=k1·∆S ’ 1+k2·∆S ’ 2+...+k k ·∆S ’ k +b, where k1~k k is the weighting coefficient for each filament, reflecting the sensitivity of the filament's jump offset to the pitch deviation, and b is a constant term; Historical data were fitted using the least squares method to determine k1~k k And b, finally we get the mapping model of jump offset - pitch deviation; For current monitoring data identified as true pitch anomalies, extract the jump value offset of all filaments and calculate the contribution of each filament to the pitch deviation. The formula is: Contribution. i =|k i |×|∆S i |, where k i It is the weight coefficient of the i-th monofilament in the historical multiple linear regression model, |∆S i | is the absolute value of the jump value offset of the i-th monofilament in the current monitoring; The contribution of each monofilament is compared with the preset contribution, and the monofilaments with a contribution greater than the preset contribution are marked as the main contributing monofilaments. Substituting the jump value offset of the main contributing monofilament into the multiple linear regression model, we obtain the current pitch deviation value, ∆H. 预测 =k1·∆S ’ 1+k2·∆S ’ 2+...+k k ·∆S ’ k +b; The current pitch deviation value is obtained by adding the pitch deviation value to the standard pitch, where the standard pitch is obtained based on the stranding machine design parameters; Stranded wire anomaly early warning module: Based on the abnormal pitch value corresponding to the stranded wire anomaly under the same working conditions, the critical value of pitch anomaly is determined, and combined with the mapping model of jump offset-pitch deviation and the actual pitch anomaly value, the remaining time of stranded wire anomaly is predicted.

[0024] The stranding abnormality refers to the continuous occurrence of abnormal actual pitch in the stranding machine. Due to the continuous deviation of the single wire spiral path (such as abnormal spiral angle caused by excessively large / small pitch), the relative position of the single wire and the sensor, and the force relationship between the single wires will gradually deteriorate, which may eventually lead to problems such as excessive wear or breakage of single wires, loose stranding or untwisted strands, serious deterioration of mechanical properties, and abnormal wear of equipment components. The process for determining the critical value of the pitch anomaly includes: Obtain the abnormal pitch value corresponding to the abnormal stranded wire caused by the abnormal pitch value under the same historical working conditions, and record the specific time when the abnormality occurred; The pitch values ​​corresponding to all abnormal moments of stranded wires were selected from historical data to form an abnormal sample set; Calculate the coefficient of variation of the abnormal sample set and compare it with the preset coefficient of variation. If the coefficient of variation is less than the preset coefficient of variation, the pitch outlier is stable; otherwise, it is unstable. If the pitch anomaly is stable, the mean of the anomaly sample set is calculated as the pitch anomaly threshold value corresponding to the twisted wire anomaly. Conversely, if it is unstable, the minimum value in the anomaly sample set is selected as the pitch anomaly threshold value corresponding to the twisted wire anomaly. The process for predicting the remaining time of the stranded wire abnormality includes: Obtain the real-time jump value offset of each monitored filament, as well as the time-series data of the jump value offset within the historical period {∆S}. ’ i (t)}; Calculate the difference ∆H between the current true pitch anomaly value and the pitch anomaly threshold value; From historical periodic monitoring data, extract the jump value offset sequence of the i-th monofilament: {∆S ’ i (t1), ∆S ’ i (t2), ...,∆S ’ i (t n )}, where t1 <t2<...<t n , is the timestamp, ∆S ’ i (t n ) represents the single filament i at timestamp t n The offset of the corresponding jump value; By linearly fitting the change of the jump value offset sequence over time, the growth rate of the jump value offset of the i-th filament is obtained. The unit is the jump value unit / s, specifically: Let the time variable be t. Starting from t1, convert it to a relative time t' = t - t1, then the fitted model is: ,in, c represents the growth rate of the offset of the jump value of the i-th monofilament. i For constant terms; Solve using the least squares method ,Right now: ; According to the jump offset-pitch deviation mapping model, the growth rate of the pitch anomaly is the weighted sum of the growth rates of the jump offsets of each monofilament, with the weights being the coefficients k in the mapping model. i, ,in, The rate of increase of the jump value offset of the i-th monofilament; The remaining time for the current true pitch anomaly to reach the pitch anomaly threshold is Where ∆H is the difference between the current true pitch anomaly value and the anomaly threshold value, v H This represents the growth rate of pitch anomalies.

[0025] The foregoing has shown and described the basic principles, main features, and advantages of the present invention. Those skilled in the art should understand that the present invention is not limited to the above embodiments. The embodiments and descriptions in the specification are merely illustrative of the principles of the invention. Various changes and modifications can be made to the invention without departing from its spirit and scope, and all such changes and modifications fall within the scope of the present invention as claimed. The scope of protection of the present invention is defined by the appended claims and their equivalents.

Claims

1. A method for monitoring the stranded wires of a stranding machine, characterized in that: include: Step 1: Perform time interval stability analysis on the jump points in the current stranding machine sensor signal, determine the jump period, and match it with the theoretical period to determine whether the current pitch measurement signal has periodic jumps; Step 2: If this occurs, determine the filament number and blocking time of the blocking sensor based on the current stranding speed of the stranding machine and the arrangement pattern of the filaments. Step 3: Within the current stranding cycle, compare the actual and theoretical occlusion and occlusion intervals of the single filaments based on the filament number and occlusion time. Combine this with the comparison and analysis of the actual jump value of the single filament with the normal range of jump values ​​to determine the coordinated characteristics of the single filaments in the jump and time dimensions, and determine whether the periodic jump of the current pitch measurement signal is caused by an abnormality in the actual pitch. Step 4: If so, establish a mapping model of jump offset-pitch deviation based on historical twisting data, and calculate the true pitch anomaly value of the filament at the current moment by combining the jump value offset of the current filament. Step 5: Determine the critical value of pitch abnormality based on the abnormal pitch value corresponding to the abnormality of the stranded wire under the same working conditions, and predict the remaining time for the abnormality of the stranded wire by combining the mapping model of jump offset-pitch deviation and the actual pitch abnormal value.

2. The strand monitoring method for a stranding machine according to claim 1, characterized in that: The method for determining whether the current pitch measurement signal exhibits periodic jumps is as follows: Calculate the first-order difference sequence of the current sensor's measurement signal: ∆x i =|x i -x i-1 |(i≥2),∆x i x represents the signal change between two adjacent points. i This represents the value of the i-th measured signal; The x corresponding to the element in the first-order difference sequence that is greater than the transition threshold. i Mark the transition point and record the transition time; All transition points are integrated into a transition point sequence, and the standard deviation of the time interval between all adjacent transition points is calculated. If the standard deviation is less than the preset standard deviation, it is preliminarily judged that there is periodicity. Calculate the theoretical monofilament blocking period of the stranding machine: T 理论 =60 / (n·k), where n is the twisting speed and k is the number of single filaments. The average time interval between adjacent jump points is calculated and used as the actual jump period of the current pitch measurement signal. The deviation between the theoretical monofilament occlusion period and the actual jump period is calculated. If the deviation is less than the preset deviation, the current pitch measurement signal will show a periodic jump.

3. The method for monitoring stranded wires in a stranding machine according to claim 1, characterized in that: The method for determining the single filament number of the occlusion sensor and the corresponding occlusion time is as follows: Obtain the current winding speed and the total number of filaments k involved in winding, and determine the ascending order of the filament numbers according to the direction of winding. The 360 ​​degrees of the winch rotation are divided into k equidistant intervals, each interval having an angle of 360° / k, and each interval corresponds to the shading period of a single filament. Calculate the current angle interval based on the rotation direction and the current angle θ, and determine the monofilament number based on the angle interval: Where k is the total number of monofilaments; The current time is t0, and the absolute angle output by the encoder in real time is θ(t0). The angle difference between the current angle and the starting point of the single-filament i-shaped occlusion interval is ∆θ=θ. start,i -θ(t0); The time difference corresponding to the angular difference is ∆t=(∆θ / 360°)·T 周期 If t is the time required to rotate by an angle ∆θ, then the occlusion start time of monofilament i is: t start,i =t0+∆t, ending time is t end,i =t start,i +T 遮挡,i .

4. The strand monitoring method for a stranding machine according to claim 1, characterized in that: The method for determining whether the periodic jumps in the current pitch measurement signal are caused by an abnormality in the actual pitch is as follows: For each monofilament involved in the twisting, suspected abnormal monofilaments are identified by comparing the actual jump value of the monofilament with the normal range of jump values. For each suspected abnormal monofilament; If the actual jump value of a single filament is greater than the upper limit of the normal jump value range, then the offset direction of the jump value of the single filament is marked as too large. If the actual jump value of a single filament is less than the lower limit of the normal jump value range, then the direction of the jump value offset of the single filament is marked as too small; If the jump values ​​of all suspected abnormal monofilaments are either too large or too small, then the jump values ​​are determined to be in the same direction. Calculate the deviation between the blocking time and the theoretical blocking time in the current twisting cycle, as well as the deviation between the blocking interval time of adjacent monofilaments and the theoretical interval time; If the deviation in duration of occlusion and the deviation in interval between adjacent monofilament occlusions are both greater than the preset deviation, and the deviations are in the same direction, either too long or too short, then the time offset direction is determined to be consistent. Suspected abnormal monofilaments with the same jump value offset direction and the same time offset direction were screened out, and the deviation of jump value and time was verified collaboratively. If the direction of the jump value offset is relatively large and the direction of the time offset is relatively long, or the direction of the jump value offset is relatively small and the direction of the time offset is relatively short, then the direction of the jump value offset and the direction of the time offset are coordinated, and the monofilament is marked as a coordinated monofilament. The proportion of cooperating monofilaments among all monofilaments is calculated to obtain the proportion of cooperating monofilaments. If the proportion of cooperating monofilaments is greater than the preset proportion, it is determined that the periodic jump in the current pitch measurement signal is caused by an abnormality in the actual pitch.

5. The strand monitoring method for a stranding machine according to claim 4, characterized in that: The method for identifying suspected abnormal transitions is as follows: Based on the 3σ principle, the upper and lower limits of the normal jump value of each single wire under the working conditions of stable operation of the stranding machine and no abnormal pitch are calculated to obtain the normal range of the jump value of the single wire. Before the single filament is blocked, the sensor is in an unblocked gap period. During the gap period, the sensor signal is collected and the average value is calculated as the stable signal value before the single filament is blocked. When collecting the signal peak value of the single-filament occlusion sensor, if it is a laser sensor, the signal peak value is the minimum value within the signal range; if it is an optical reflection sensor, the signal peak value is the maximum value within the signal range. Calculate the actual jump value ∆S of a single filament i =|S 峰，i -S 稳定 |, where S 峰，i S represents the peak signal value when a single filament i blocks the sensor. 稳定 The stable signal value before the sensor is blocked by a single filament; If the actual jump value of a single filament is ∆S i If the jump value of a single filament is outside the normal range, it is marked as a suspected abnormal jump.

6. The method for monitoring stranded wires in a stranding machine according to claim 1, characterized in that: The construction process of the mapping model of jump offset-pitch deviation is as follows: Under known pitch deviation conditions, historical winding data are recorded synchronously, including: pitch deviation ∆H. 历史 =H 历史 -H0, where H0 is the standard pitch, H 历史 The actual pitch measured offline, and the offset ∆S of the jump value of each monofilament. 历史,i , that is, the deviation between the jump value of the i-th monofilament under this historical working condition and the normal jump average value; The multiple linear regression model is adopted, and the expression is ∆H=k1·∆S ’ 1+k2·∆S ’ 2+...+k k ·∆S ’ k +b, where k1~k k Here, is the weighting coefficient for each monofilament, and b is a constant term; The historical twisting data was fitted using the least squares method to determine k1~k k And b, finally we get the mapping model of jump offset - pitch deviation.

7. The strand monitoring method for a stranding machine according to claim 6, characterized in that: The calculation method for the true pitch anomaly value is as follows: Obtain the jump value offset of all single filaments, and calculate the contribution of each single filament to the pitch deviation. The formula is: Contribution i =|k i |×|∆S i |, where k i It is the weight coefficient of the i-th monofilament in the multiple linear regression model, |∆S i | is the absolute value of the jump value offset of the i-th monofilament in the current monitoring; Single filaments with a contribution greater than the preset contribution are marked as major contributing single filaments; Substituting the jump value offset of the main contributing monofilament into the multiple linear regression model, the current pitch deviation value is obtained; The current pitch deviation value is added to the standard pitch to obtain the current true pitch anomaly value.

8. The strand monitoring method for a stranding machine according to claim 7, characterized in that: The method for determining the critical value of the pitch anomaly is as follows: Obtain the abnormal pitch values ​​of all stranded wire abnormalities during historical stranding processes to form an abnormal sample set; Calculate the coefficient of variation of the abnormal sample set. If the coefficient of variation is less than the preset coefficient of variation, calculate the mean of the abnormal sample set as the critical value of the pitch abnormality corresponding to the twisted wire abnormality. Otherwise, select the minimum value in the abnormal sample set as the critical value of the pitch abnormality corresponding to the twisted wire abnormality.

9. The method for monitoring stranded wires in a stranding machine according to claim 8, characterized in that: The prediction process for the remaining time of the stranded wire abnormality is as follows: Calculate the difference ∆H between the current true pitch anomaly value and the pitch anomaly threshold value; Extract the jump value offset ∆S of each monofilament from historical twisting data. 历史,i The jump value offset sequence is obtained: {∆S ’ i (t1), ∆S ’ i (t2), ...,∆S ’ i (t n )}, where t1 <t2<...<t n , is the timestamp, ∆S ’ i (t n ) represents the single filament i at timestamp t n The offset of the corresponding jump value; By linearly fitting the change in the jump value offset sequence over time, a fitting model is obtained: ,in, c represents the growth rate of the offset of the jump value of the i-th monofilament. i For constant terms; The growth rate of the jump value offset of the i-th monofilament is obtained by solving the least squares method. ; The growth rate of the pitch outlier is the weighted sum of the growth rates of the jump values ​​of each monofilament, with the weights being the coefficients k in the jump offset-pitch deviation mapping model. i The specific weighting formula is as follows: ,in, v represents the rate of increase of the jump value offset of the i-th monofilament. H This represents the growth rate of the pitch anomaly. The remaining time for the current true pitch anomaly to reach the pitch anomaly threshold is , where ∆H is the difference between the current true pitch anomaly value and the anomaly threshold value.

10. A strand monitoring system for a stranding machine, comprising the following modules: Periodic jump identification module: Performs time interval stability analysis on the jump points in the current stranding machine sensor signal, determines the jump period, and performs matching analysis with the theoretical period to determine whether the current pitch measurement signal has a periodic jump; Single filament obstruction determination module: If obstruction occurs, the single filament number and obstruction time of the obstruction sensor are determined based on the current stranding speed of the stranding machine and the single filament arrangement pattern. Actual pitch anomaly judgment module: Within the current stranding cycle, the module compares the actual and theoretical occlusion and occlusion interval of the single filament based on the single filament number and occlusion time, and combines the comparison and analysis of the actual jump value of the single filament with the normal range of jump value to determine the coordinated characteristics of the single filament in the jump and time dimensions, and judges whether the periodic jump of the current pitch measurement signal is caused by an actual pitch anomaly. Pitch anomaly calculation module: If so, establish a mapping model of jump offset - pitch deviation based on historical twisting data, and calculate the real pitch anomaly of the filament at the current moment by combining the jump value offset of the current filament. Stranded wire anomaly early warning module: Based on the abnormal pitch value corresponding to the stranded wire anomaly under the same working conditions, the critical value of pitch anomaly is determined, and combined with the mapping model of jump offset-pitch deviation and the actual pitch anomaly value, the remaining time of stranded wire anomaly is predicted.

Images