A method and system for on-line monitoring of a spinning frame single spindle

Abstract

The application relates to the spinning technology field, in particular to a spinning frame single spindle online monitoring method and system, and specifically comprises the following: the application analyzes the shadow shielding generated in the patrol process of a spinning frame automatic joint inspection robot in a workshop during operation of the spinning frame, and the influence of photoelectric sensor data collection on single spindle monitoring when flying fluff appears in the workshop, constructs an abnormality discrimination value of each abnormal spindle, determines the real abnormal spindle based on the abnormality discrimination value, and controls a roving stop feeding device to stop feeding roving, so that the problem that the traditional method for monitoring the single spindle of the spinning frame is prone to being affected by the shadow shielding of the spinning frame automatic joint inspection robot and the low photoelectric sensor data collection accuracy when flying fluff appears in the workshop, and the false alarm of abnormal spindle judgment is caused is solved, the identification accuracy of the abnormal spindle is improved, and the accuracy of the single spindle online monitoring of the spinning frame is ensured.

Description

TECHNICAL FIELD

[0001] The present application relates to the technical field of spinning, in particular to a fine spinning machine single spindle online monitoring method and system. BACKGROUND

[0002] In the process of fine spinning, the high-speed rotation of the ring traveler makes the yarn in the spinning section have a large tension, so that the fine yarn is prone to breakage. At the same time, due to the existence of the twisting triangle area in ring spinning, the edge fibers in the triangle area bear a larger tension, which also causes the fine yarn to break. Again, due to the blocking effect in the process of twist transmission from bottom to top, the twist of the spinning section is reduced, thereby producing a weak twist section, which causes the fine yarn to break. Therefore, online detection of breakage in ring spinning is currently a hot research topic.

[0003] Currently, fine spinning single spindle detection mainly uses photoelectric sensors to monitor the operation of the spindle. Based on the reflection characteristics of light and the feature that the spindle speed will change when the spindle breaks or has a weak twist anomaly, the rotation speed is calculated by detecting the periodic change of the light signal during the rotation of the ring traveler, so as to determine whether there is a breakage, a weak twist anomaly, or the like. However, when using a photoelectric sensor to measure the speed, it is easily affected by factors such as background color and external environment. For example, changes in temperature and humidity in the workshop can cause condensation on the surface of the photoelectric sensor, affecting detection accuracy. Flying debris can block the light signal received by the sensor, causing the received light signal intensity to decrease. The color of the ring traveler and the ring is similar, which may cause false positives. In addition, when the fine yarn automatic joint inspection robot passes through the production workshop, the shadow of the fine yarn automatic joint inspection robot caused by the light will block the photoelectric sensor, thereby interfering with the light signal intensity collected by the photoelectric sensor, and the accuracy of the measured rotation speed data will decrease, ultimately misjudging the actual working condition of the spindle, thereby affecting production efficiency and quality. SUMMARY

[0004] In order to solve the above technical problems, the purpose of the present application is to provide a fine spinning machine single spindle online monitoring method and system, and the technical solution adopted is as follows:

[0005] In the first aspect, the present application provides a fine spinning machine single spindle online monitoring method, which comprises the following steps:

[0006] Collect the ring traveler rotation speed of each spindle of the fine spinning machine at each time, and construct a rotation speed data sequence of each spindle;

[0007] The spindle with a mutation point in the rotation speed data sequence is regarded as an abnormal spindle. Among the spindles in the same row of the fine spinning machine, based on the similarity between the rotation speed change characteristics of each abnormal spindle and the abnormal spindles on its left and right sides, and combining the difference between the rotation speed mutation time of each abnormal spindle and the remaining abnormal spindles, the true abnormality degree of each abnormal spindle is constructed.

[0008] a change judgment sequence of the abnormal spindle is formed by the data between the maximum points in the rotation speed data sequence of the abnormal spindle;

[0009] a flying interference coefficient of each abnormal spindle is constructed based on the data clustering distribution characteristics in each change judgment sequence and the slope change characteristics on the fitting curve of each change judgment sequence;

[0010] an abnormal discrimination value of each abnormal spindle is constructed based on the real abnormal degree and the flying interference coefficient; a real abnormal spindle is determined based on the abnormal discrimination value, and the on-line monitoring of the spinning frame single spindle is performed.

[0011] In one embodiment, the acquisition process of the real abnormal degree is as follows:

[0012] the maximum mutation point in the rotation speed data sequence is taken as the rotation speed mutation time of the corresponding abnormal spindle;

[0013] a left rotation speed change amount sequence and a right rotation speed change amount sequence of any abnormal spindle are respectively constructed based on the element range in the rotation speed data sequence of each abnormal spindle;

[0014] a real abnormal degree of each abnormal spindle is constructed based on the difference between the rotation speed mutation time of each abnormal spindle and that of other abnormal spindles, and the variance of the first derivative sequence of the left rotation speed change amount sequence and the right rotation speed change amount sequence.

[0015] In one embodiment, the left rotation speed change amount sequence and the right rotation speed change amount sequence are as follows:

[0016] the sequence composed of the element range of the left abnormal spindle in the same row of the i-th abnormal spindle is taken as the left rotation speed change amount sequence, and the sequence composed of the element range of the right abnormal spindle is taken as the right rotation speed change amount sequence.

[0017] In one embodiment, the expression of the real abnormal degree is as follows:

[0018]

[0019] In the formula, is the real abnormal degree of the i-th abnormal spindle; is the mean value of the absolute value of the difference between the rotation speed mutation time of the i-th abnormal spindle and that of other abnormal spindles; is the variance of the first derivative sequence of the left rotation speed change amount sequence of the i-th abnormal spindle; is the variance of the first derivative sequence of the right rotation speed change amount sequence of the i-th abnormal spindle; is a normalization function.

[0020] In one embodiment, the acquisition process of the fly-fluff interference coefficient is as follows:

[0021] Determine the change judgment sequence of the corresponding abnormal spindle based on the position of the maximum value in the rotation speed data sequence;

[0022] Cluster all data in the change judgment sequence to obtain each cluster; fit the change judgment sequence by a fitting algorithm to obtain a fitting curve, and obtain the slope of the connecting line between adjacent data points on the fitting curve;

[0023] Calculate the normalized value of the variance of all mutation points in the rotation speed data sequence of all abnormal spindles; based on the element quantity distribution characteristics in the cluster and the data distribution characteristics of the slope of the connecting line on the fitting curve, and in combination with the normalized value of the variance of all mutation points, construct the fly-fluff interference coefficient of the abnormal spindle.

[0024] In one embodiment, the change judgment sequence is specifically as follows: the sequence composed of the rotation speed data between the maximum value point and the minimum value point in the rotation speed data sequence is taken as the change judgment sequence of the corresponding abnormal spindle.

[0025] In one embodiment, the construction of the fly-fluff interference coefficient of the abnormal spindle is specifically as follows:

[0026] Calculate the mean value of the normalized value of the number of elements in all clusters in the change judgment sequence, denoted as the first mean value; among the slopes of all connecting lines on the fitting curve of the change judgment sequence, calculate the mean value of the normalized value of the slopes with positive values, denoted as the second mean value;

[0027] Take the sum of the first mean value, the second mean value and the normalized value of the variance of all mutation points as the fly-fluff interference coefficient of the corresponding abnormal spindle.

[0028] In one embodiment, the abnormal discrimination value is the ratio of the true abnormal degree of each abnormal spindle to the fly-fluff interference coefficient.

[0029] In one embodiment, the determination of the real abnormal spindle based on the abnormal discrimination value is specifically as follows:

[0030] Obtain a segmentation threshold value by a threshold segmentation algorithm for the abnormal discrimination values of all abnormal spindles; take the spindle with an abnormal discrimination value greater than or equal to the segmentation threshold value as a real abnormal spindle, and otherwise as a false abnormal spindle;

[0031] When a real abnormal spindle is detected, perform self-stop control and digital display according to the number and position information of the real abnormal spindle.

[0032] In a second aspect, the embodiments of the present application also provide a spinning frame single spindle online monitoring system, comprising a memory, a processor, and a computer program stored in the memory and running on the processor, and the processor implements the steps of the method of any one of the above when executing the computer program.

[0033] The embodiments of the present application have at least the following beneficial effects:

[0034] The present application analyzes the shadow blocking in the patrol process of the spinning automatic joint inspection robot in the spinning frame running process and the influence of the photoelectric sensor data collection of the single spindle monitoring when the flying fluff appears in the workshop, deduces a judgment index for measuring whether the abnormal condition of a single spindle is real, and then the spinning frame single spindle online monitoring roving self-stop system can more accurately judge the real degree of the abnormal condition of the abnormal spindle, thereby determining the real abnormal spindle and controlling the roving stop feeding device to stop feeding the roving, solving the problem that the current spinning frame single spindle monitoring by using the traditional method is easily affected by the shadow blocking of the spinning automatic joint inspection robot and the low data collection accuracy of the photoelectric sensor when the flying fluff appears in the workshop, leading to false positives in the abnormal judgment of the spindle, improving the detection accuracy of the abnormal spindle, and ensuring the accuracy of the spinning frame single spindle online monitoring. BRIEF DESCRIPTION OF DRAWINGS

[0035] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present application or the prior art, the following will briefly introduce the drawings needed to be used in the embodiments or prior art description. Obviously, the drawings in the following description are only some embodiments of the present application, and those skilled in the art can obtain other drawings according to these drawings without any creative effort.

[0036] Figure 1 A step flowchart of a spinning frame single spindle online monitoring method provided by an embodiment of the present application;

[0037] Figure 2 A schematic diagram of the real abnormality degree acquisition process. DETAILED DESCRIPTION

[0038] In order to further illustrate the technical means and effects adopted by the present application to achieve the predetermined invention purposes, the following describes the specific embodiments, structure, features and effects of the spinning frame single spindle online monitoring method and system according to the present application in combination with the drawings and preferred embodiments. In the following description, different "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. In addition, the specific features, structures or characteristics in one or more embodiments can be combined in any suitable form.

[0039] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this application belongs.

[0040] The specific scheme of the method and system for monitoring a spinning frame single spindle on line is described in detail below with reference to the drawings.

[0041] Referring to Figure 1 which shows a flow chart of steps of a method for monitoring a spinning frame single spindle on line according to an embodiment of the application, the method comprises the following steps:

[0042] In step S1, the rotating speed of the ring of each spindle of the spinning frame at each time is collected, and a rotating speed data sequence of each spindle is constructed.

[0043] A workshop comprises a plurality of spinning frames, each of which comprises two rows of spindles, and the spindles are usually arranged in equal intervals in a horizontal direction. The rotating speed data of each spindle of the spinning frame is acquired every interval T through a system for stopping the spinning frame on line, and all the rotating speed data of each spindle is arranged in ascending order of time to form a sequence, which is referred to as the rotating speed data sequence of each spindle.

[0044] Preferably, in the embodiment of the application, the value of T is set to 0.1s. As other embodiments of the application, the implementer can set the value of T according to the actual situation.

[0045] In order to facilitate subsequent processing, the collected rotating speed data sequence is normalized.

[0046] In step S2, the spindles with mutation points in the rotating speed data sequence are regarded as abnormal spindles, and the real abnormal degree of each abnormal spindle is constructed based on the similarity between the rotating speed change characteristics of each abnormal spindle and the abnormal spindle on each side thereof and the difference between the rotating speed mutation time of each abnormal spindle and the other abnormal spindles in the same row of the spinning frame.

[0047] A spinning frame is a machine device in the textile industry for spinning roving or sliver into yarn. The spinning frame mainly consists of a feeding mechanism, a drafting mechanism, a twisting and winding mechanism, etc. Its function is to draft, twist and wind the roving or sliver sent from the previous process through a series of processes, and finally produce yarn products that meet the requirements. At present, full-automatic production equipment is usually adopted. When the speed exceeds or is lower than the preset threshold value, it is considered that the spindle has an abnormal situation such as broken end or weak twist, and then the spinning automatic joint inspection robot needs to go to the specified area for abnormal treatment. Therefore, in the daily production process, there are multiple spinning automatic joint inspection robots in the workshop for inspection or spinning joint for abnormal spindles. In the process of robot inspection, the shadow area generated by the spinning automatic joint inspection robot passing through the spinning frame may cover the position of the photoelectric sensor for monitoring the speed of the steel ring of the spindle, thereby interfering with the light signal collected by the photoelectric sensor, and affecting the accuracy of the speed data. Specifically, when the shadow is blocked, the speed data decreases. If the blockage is serious, the speed may be lower than the threshold value. At this time, the spinning frame single-spindle monitoring module may judge this situation as an abnormality, thereby causing a misjudgment.

[0048] For the shadow interference generated by the spinning automatic joint inspection robot, since the shadow area is relatively large, the number of spindles affected is not unique. Therefore, when a spindle is affected by the shadow, its adjacent spindles may also be affected to a certain extent. Because the spindles of the spinning frame are arranged in a horizontal row at equal intervals, and the brightness of the central area of the shadow is usually lower, the farther away from the central area of the shadow, the higher the brightness. Based on this, when a certain spindle is affected by the shadow, the degree of influence of the left and right adjacent spindles decreases obviously. Specifically, the speed data of the left and right adjacent spindles of a single spindle decreases by a similar degree.

[0049] In addition, when the spinning automatic joint inspection robot passes through the spindles of the spinning frame, the shadow appears instantaneously and covers the spindles. Therefore, when the spinning automatic joint inspection robot shadow interference occurs, the time when the affected spindles change speed is relatively close. Specifically, the mutation points of the speed data correspond to a small difference in time.

[0050] (1) Taking any one spindle in a single spinning frame as an example, the speed data sequence of the spindle is taken as the input of the mutation point detection algorithm, and the output is the mutation point in the speed data sequence of the spindle. The mutation point detection algorithm is a known technology, and the specific process is not described again.

[0051] Preferably, in the embodiments of the present application, the Mann-Kendall (MK) test algorithm is used to detect the mutation points in the rotation speed data sequence. As other embodiments of the present application, the implementer can also use other mutation point detection algorithms such as Pettitt and CUSUM (cumulative sum control chart), and the present application does not make specific limitations.

[0052] It should be noted that since there may be a case where the spinning frame is completely normal, at this time the rotation speed data in the spindle is constant, so that the mutation point cannot be detected, at this time the spindle is not processed.

[0053] Further, when there is a mutation point in the rotation speed data sequence of the spindle, the spindle is taken as an abnormal spindle. The time corresponding to the maximum mutation point in the rotation speed data sequence of each abnormal spindle is taken as the rotation speed mutation time of the abnormal spindle; taking the i th abnormal spindle as an example, the difference between the rotation speed mutation time of the i th abnormal spindle and that of each abnormal spindle is calculated, which is denoted as the first difference absolute value.

[0054] (2) The range of all elements in the rotation speed data sequence of each abnormal spindle is calculated, which is denoted as the rotation speed range of the abnormal spindle; further, for the i th abnormal spindle in the spinning frame, the abnormal spindle on the left side of the abnormal spindle in the row where the abnormal spindle is located is taken as the left abnormal spindle of the i th abnormal spindle; the rotation speed range of all left abnormal spindles of the i th abnormal spindle is arranged in order from near to far according to the distance from the left abnormal spindle to the i th abnormal spindle, and the sequence is taken as the left rotation speed change amount sequence of the i th abnormal spindle.

[0055] For example, there are 5 abnormal spindles on the left side of the i th abnormal spindle, and the intervals between the i th abnormal spindle and the 5 abnormal spindles are 5, 4, 3, 2 and 1 respectively, and the rotation speed data sequence ranges of the 5 abnormal spindles are 20, 18, 16, 14 and 12 respectively, then the left rotation speed change amount sequence of the i th abnormal spindle is [12, 14, 16, 18, 20].

[0056] Similarly, the abnormal spindle on the right side of the i th abnormal spindle in the row where the abnormal spindle is located is taken as the right abnormal spindle of the i th abnormal spindle; the rotation speed range of all right abnormal spindles of the i th abnormal spindle is arranged in order from near to far according to the distance from the right abnormal spindle to the i th abnormal spindle, and the sequence is taken as the right rotation speed change amount sequence of the i th abnormal spindle.

[0057] (3) In order to judge the authenticity of the rotation speed drop of a single abnormal spindle, based on the above analysis, the true abnormal degree of each abnormal spindle is constructed, and the expression is:

[0058]

[0059] In the formula, the true abnormal degree of the i th abnormal spindle is denoted as TAD i, the first difference absolute value of the i th abnormal spindle is denoted as DAD i, the rotation speed range of the i th abnormal spindle is denoted as R i, and the rotation speed range of the j th abnormal spindle is denoted as R j. the real abnormal degree of the ith abnormal spindle; the mean value of all the first difference absolute values of the ith abnormal spindle; the variance of the first derivative sequence of the left speed change amount sequence of the ith abnormal spindle; the variance of the first derivative sequence of the right speed change amount sequence of the ith abnormal spindle; is a normalization function. The first derivative is a known technology, and the specific process will not be repeated.

[0060] It should be noted that if there is no abnormal spindle on the left and right sides of the ith abnormal spindle, is 0. In addition, only when there is no other abnormal spindle in the row where the ith abnormal spindle is located, .

[0061] When an abnormal spindle appears in a row of the spinning machine, if there are other abnormal spindles in the row, and the time when the speed of the abnormal spindles changes is closer, and the change degree of the speed change amount on the left and right sides of the single abnormal spindle is closer, it means that the abnormal spindle is more likely to be caused by the shadow blocking of the automatic splicing robot, and the real degree of the abnormality is lower. On the contrary, the possibility of the spindle being a real abnormality is higher. In particular, if there is only one spindle in a row of the spinning machine that is abnormal, the possibility of it being a real abnormality is extremely high.

[0062] Step S3, forming a change judgment sequence of the abnormal spindle through the data between the extreme points in the speed data sequence of the abnormal spindle; based on the data clustering distribution characteristics in each change judgment sequence, and the slope change characteristics on the fitting curve of each change judgment sequence, a fly interference coefficient of each abnormal spindle is constructed.

[0063] In the spinning machine production workshop, fly is usually referred to as the phenomenon that fibers fall off from the sliver or roving due to various reasons in the spinning process and float in the air. Due to the different degrees of fly, the interference degree of each spindle is different, and due to the influence of the ventilation system in the workshop, the distribution of the spindles affected by the fly is relatively discrete. Therefore, when only the above method is used for calculation, the spindles affected by the fly may be judged as real abnormal spindles, and then misjudgment may occur, so further analysis is needed.

[0064] When the flying debris occurs, the flying debris can block the data collection of the photoelectric sensor. When the flying debris is less dense and the blocking is weak, the photoelectric sensor is less disturbed, which is specifically manifested in that the collected rotation speed data appears irregular fluctuations, and the fluctuation intensity difference is large. When the flying debris is more dense, the flying debris can be attached to the photoelectric sensor, thereby greatly interfering with the data collection of the photoelectric sensor, and finally causing the collected rotation speed data to appear a large drop. In addition, the flying debris is affected by the ventilation system in the workshop, the flying debris has a low movement regularity and is unevenly distributed, the positions of the spindles disturbed by the flying debris are relatively discrete, and the disturbance degree is different. Specifically, in the rotation speed data, part of the spindle data appears fluctuations, and the duration, amplitude and occurrence time of each fluctuation are different.

[0065] For the case that the flying debris is gathered in a certain photoelectric sensor, the disturbance of the photoelectric sensor is gradually increased over time due to the accumulation of the flying debris. Compared with the real weak twist abnormality, the rotation speed data measured by the photoelectric sensor disturbed by the flying debris has a low smoothness and speed of change, and due to the influence of the ventilation system in the workshop, the flying debris may decrease during the gathering process in the photoelectric sensor, which is manifested in that the rotation speed data decreases gradually and reverses during the decrease process.

[0066] (1) In order to represent the influence of the flying debris on the rotation speed data collected by the photoelectric sensor, taking the rotation speed data sequence of a single abnormal spindle as an example, the maximum point and the minimum point in the rotation speed data sequence of the current abnormal spindle are obtained, and the sequence composed of the rotation speed data between the maximum point and the minimum point in the rotation speed data sequence is taken as the change judgment sequence of the current abnormal spindle.

[0067] (2) All data in the change judgment sequence are taken as the input of the DPC (Density Peak Clustering Algorithm) density clustering algorithm, wherein the cutoff distance is set to 1, and the output is a plurality of clustering clusters, and the size of all data in each clustering cluster is the same. The DPC density clustering algorithm is a known technology, and the specific process is not described again. It should be noted that the DPC clustering algorithm is mainly used to obtain the data with the same size and length in the change judgment sequence, so as to distinguish the data change characteristics of the flying debris and the weak twist.

[0068] It should be noted that for the clustering of the data in the change judgment sequence, the present application only provides one clustering method, and there are many existing clustering methods, and the implementer can also use other clustering algorithms to cluster the data in the change judgment sequence, and the present application does not make specific limitations.

[0069] If multiple data are contained in a single cluster, it indicates that multiple same speed data exist in a certain time in the variation judgment sequence, i.e. a stable stage of speed exists.

[0070] Further, the number of elements in each cluster is counted, and the number of elements in all clusters is normalized. In the slope of all lines in the fitting curve of the variation judgment sequence, the average of the normalized values of the slopes with positive values is calculated, and is recorded as a first average.

[0071] (3) The variation judgment sequence is taken as an input of a least square curve fitting algorithm to perform curve fitting, and a fitting curve of the variation judgment sequence is obtained; the least square curve fitting algorithm is a known technology, and a specific process is not described herein.

[0072] It should be noted that for curve fitting of the variation judgment sequence, the present application only provides a curve fitting method, and there are many existing curve fitting methods, and implementers can also use other clustering algorithms to perform curve fitting on the variation judgment slope, which is not specifically limited in the present application.

[0073] Further, the slope of the line between each data point and the corresponding previous data point in the fitting curve except the first data point is calculated, and all calculated slope data are normalized. The average of the normalized values of the number of elements in all clusters in the variation judgment sequence is calculated, and is recorded as a second average.

[0074] If the current abnormal spindle is a real abnormal situation, the monotonicity of the slope of the data point line in the spindle variation judgment sequence curve is consistent, and the degree of slope change is approximately. When the abnormal situation of a single spindle is caused by flying hair, the smoothness of the variation judgment sequence curve is low, and the turning back situation may occur, i.e. a large positive slope may exist in the variation judgment sequence curve.

[0075] (4) To further judge the authenticity of the abnormal situation of a single abnormal spindle, based on the above analysis, a flying hair interference coefficient of each abnormal spindle is constructed, and the expression is:

[0076]

[0077] In the formula, is the flying hair interference coefficient of the i-th abnormal spindle is the first average of the variation judgment sequence of the i-th abnormal spindle is the normalized value of the variance of all mutation points in the speed data sequence of all abnormal spindles is the second average of the variation judgment sequence of the i-th abnormal spindle.

[0078] It should be noted that if there is no positive slope in the fitting curve of the change judgment sequence of the i-th abnormal spindle, the average of the total number of data in all cluster clusters in the change judgment sequence of the i-th abnormal spindle is set to 0. is set to 0.

[0079] When the average of the total number of data in all cluster clusters in the change judgment sequence of the i-th abnormal spindle is greater, the variance of the value of the speed mutation point of the remaining abnormal spindles is greater, and a positive slope appears in the change judgment sequence curve and the average is greater, it indicates that the speed data of the spindle is more likely to appear a stable stage and the fluctuation difference with the remaining abnormal spindles is more obvious. The frequency of the turning back in the change judgment sequence of the spindle is higher, and the amplitude is greater, that is, the abnormal of the spindle is more likely to be caused by the flying situation in the workshop, rather than a real abnormal.

[0080] Step S4, constructing an abnormal discrimination value of each abnormal spindle based on the real abnormal degree and the flying interference coefficient; determining the real abnormal spindle based on the abnormal discrimination value to perform the online monitoring of the spinning frame single spindle.

[0081] To evaluate the real situation of each spindle in the current spinning frame, the abnormal discrimination value of each abnormal spindle is constructed based on the real abnormal degree and the flying interference coefficient of each abnormal spindle, and the expression is:

[0082]

[0083] In the formula, is the abnormal discrimination value of the i-th abnormal spindle, is the real abnormal degree of the i-th abnormal spindle, is the flying interference coefficient of the i-th abnormal spindle.

[0084] When the real abnormal degree of the i-th abnormal spindle is greater, and the flying interference coefficient is smaller, it indicates that the possibility of the existence of the abnormal of the spindle is higher, and the authenticity is stronger.

[0085] Further, the abnormal discrimination values of all abnormal spindles are taken as input, and the output is a segmentation threshold value of the abnormal discrimination value by using cross-validation. When the abnormal discrimination value of a single abnormal spindle is greater than or equal to the segmentation threshold value, it is considered that the spindle is a real abnormal spindle, otherwise it is a false abnormal spindle.

[0086] Further, when the real abnormal spindle is detected, the spinning frame single spindle online monitoring roving self-stop system performs self-stop control and digital display according to the number and position information of the real abnormal spindle, specifically:

[0087] The roving frame single spindle online monitoring roving self-stop system sends a roving stop feeding control signal, an LED lamp flashing control signal, an LED digital screen displaying spindle fault conditions and a fault number signal. After the roving stop feeding device receives the control signal, the roving stop feeding operation is performed, the LED lamp control module corresponding to the spindle performs the LED lamp flashing control, and the LED digital screen control module digitally displays the number and position of the abnormal spindle. Thus, the roving frame single spindle online monitoring is completed. It should be noted that the self-stop control and digital display process are known technologies, and the specific process will not be described again.

[0088] The acquisition process of the real abnormality degree is shown in the schematic diagram as Figure 2

[0089] Based on the same inventive concept as the above method, the application embodiment also provides a roving frame single spindle online monitoring system, which comprises a memory, a processor, and a computer program stored in the memory and running on the processor, and the processor implements the steps of any one of the methods in the above roving frame single spindle online monitoring method when executing the computer program.

[0090] In summary, the application embodiment provides a roving frame single spindle online monitoring method. By analyzing the shadow shielding generated in the patrol process of the automatic joint inspection robot in the spinning mill and the influence of the photoelectric sensor data collection of the single spindle monitoring when the flying fluff appears in the workshop, a judgment index for measuring whether the abnormal condition of a single spindle is real is derived. Then, the roving frame single spindle online monitoring roving self-stop system can more accurately judge the real degree of the abnormal condition of the abnormal spindle, so as to determine the real abnormal spindle and control the roving stop feeding device to stop feeding the roving. The problem of false positives in the abnormal judgment of the spindle caused by the low accuracy of the photoelectric sensor data collection when the spinning mill is easily affected by the shadow shielding of the automatic joint inspection robot and the flying fluff in the workshop is solved. The detection accuracy of the abnormal spindle is improved, and the accuracy of the single spindle online monitoring of the roving frame is ensured.

[0091] It should be noted that the above-mentioned application embodiments are in the order of description only, and do not represent the advantages and disadvantages of the embodiments. The above describes the specific embodiments of the application. In addition, the processes depicted in the drawings do not necessarily require the specific order or continuous order shown to achieve the desired results. In some embodiments, multi-task processing and parallel processing are possible or may be advantageous.

[0092] Each embodiment in the application is described in a progressive manner, and the same or similar parts of each embodiment can be referred to each other. Each embodiment focuses on the differences from other embodiments.

[0093] ​The above description is only the preferred embodiment of the present application, and is not used to limit the present application, any modification, equivalent replacement, improvement, etc. made within the principle of the present application should be included in the protection scope of the present application.

Claims

1. A method of on-line monitoring of a spinning frame single spindle, characterized in that, The method comprises the following steps: Collecting the ring traveler rotating speed of each spindle of the spinning frame at each time, and constructing a rotating speed data sequence of each spindle; Taking the spindles with mutation points in the rotating speed data sequence as abnormal spindles; in the spindles in the same row of the spinning frame, based on the similarity between the rotating speed change characteristics of each abnormal spindle and the abnormal spindles on the left and right sides thereof, and combining the difference between the rotating speed mutation time of each abnormal spindle and the remaining abnormal spindles, the real abnormal degree of each abnormal spindle is constructed; Through the data between the extreme points in the rotating speed data sequence of the abnormal spindle, a change judgment sequence of the abnormal spindle is formed; based on the data clustering distribution characteristics in each change judgment sequence, and the slope change characteristics on the fitting curve of each change judgment sequence, the flying interference coefficient of each abnormal spindle is constructed; Based on the real abnormal degree and the flying interference coefficient, an abnormal discrimination value of each abnormal spindle is constructed; based on the abnormal discrimination value, the real abnormal spindles are determined, and the on-line monitoring of the single spindle of the spinning frame is performed; The acquisition process of the real abnormal degree is as follows: Taking the maximum mutation point in the rotating speed data sequence as the rotating speed mutation time of the corresponding abnormal spindle; Based on the element range in the rotating speed data sequence of each abnormal spindle, the left rotating speed change amount sequence and the right rotating speed change amount sequence of any abnormal spindle are respectively constructed; Based on the difference between the rotating speed mutation time of each abnormal spindle and other abnormal spindles, and combining the variance of the first derivative sequence of the left rotating speed change amount sequence and the right rotating speed change amount sequence, the real abnormal degree of each abnormal spindle is constructed; The left rotating speed change amount sequence and the right rotating speed change amount sequence are specifically as follows: Taking the sequence composed of the element range of the left abnormal spindle among the spindles in the same row of the ith abnormal spindle as the left rotating speed change amount sequence, and taking the sequence composed of the element range of the right abnormal spindle as the right rotating speed change amount sequence; Based on the abnormal discrimination value, the real abnormal spindles are determined, and the on-line monitoring of the single spindle of the spinning frame is performed, specifically as follows: The abnormal discrimination values of all abnormal spindles are obtained by a threshold segmentation algorithm to obtain a segmentation threshold; the spindles with the abnormal discrimination values greater than or equal to the segmentation threshold are taken as real abnormal spindles, and otherwise as false abnormal spindles; When the real abnormal spindles are detected, self-stop control and digital display are performed according to the number and position information of the real abnormal spindles.

2. A method of on-line monitoring of a single spindle of a spinning frame as claimed in claim 1, characterized in that, The expression of the real abnormal degree is as follows: In the formula, is the real abnormality degree of the ith abnormal spindle; is the mean value of the absolute value of the difference between the speed mutation time of the ith abnormal spindle and the speed mutation time of other abnormal spindles; is the variance of the first derivative sequence of the left side speed change amount sequence of the ith abnormal spindle; is the variance of the first derivative sequence of the right side speed change amount sequence of the ith abnormal spindle; is a normalization function.

3. A method of on-line monitoring of a single spindle of a spinning frame as claimed in claim 1, wherein, The acquisition process of the flying interference coefficient is as follows: Based on the extreme value position in the rotating speed data sequence, the change judgment sequence of the corresponding abnormal spindle is determined; All data in the change judgment sequence are clustered to obtain each cluster; the change judgment sequence is curve-fitted by a fitting algorithm to obtain a fitting curve, and the slope of the connecting line between adjacent data points on the fitting curve is obtained; The normalized value of the variance of all mutation points in the rotating speed data sequence of all abnormal spindles is calculated; based on the element number distribution characteristics in the cluster, and the data distribution characteristics of the slope of the connecting line on the fitting curve, and combining the normalized value of the variance of all mutation points, the flying interference coefficient of the abnormal spindle is constructed.

4. A method of on-line monitoring of a single spindle of a spinning frame as claimed in claim 3, characterized in that, The change judgment sequence is specifically as follows: the sequence composed of the rotating speed data between the maximum value point and the minimum value point in the rotating speed data sequence is taken as the change judgment sequence of the corresponding abnormal spindle.

5. A method of on-line monitoring of a single spindle of a spinning frame as claimed in claim 3, wherein, The fly interference coefficient of the constructed abnormal spool is specifically: The normalized value of the number of elements in all clustering clusters in the change judgment sequence is calculated, and the average is recorded as the first average value; among the slopes of all the lines on the fitting curve of the change judgment sequence, the normalized value of the slopes with positive values is calculated, and the average is recorded as the second average value; The sum of the first average value, the second average value and the normalized value of the variance of all the mutation points is taken as the fly interference coefficient of the corresponding abnormal spool.

6. A method of on-line monitoring of a single spindle of a spinning frame as claimed in claim 1, wherein, The abnormal discrimination value is the ratio of the true abnormal degree of each abnormal spool to the fly interference coefficient.

7. A spinning frame single spindle on-line monitoring system comprising a memory, a processor and a computer program stored in the memory and running on the processor, characterized in that, The processor implements the steps of the method of any one of claims 1-6 when executing the computer program.

Images