Tension control method and system of high-speed loom

Abstract

The invention relates to the field of textile control, in particular to a tension control method and system of a high-speed loom. The method comprises the following steps: firstly, obtaining the tension fluctuation degree of each moment according to the tension data difference of the loom at each moment and the tension data distribution of each moment in a preset time domain of each moment, clustering all moments to obtain a first cluster, and obtaining the tension fluctuation degree of each moment according to the tension data distribution of each moment in each first cluster; the method comprises the steps of obtaining a change state value of each first clustering cluster according to the tension fluctuation degree of each first clustering cluster and the difference of the tension fluctuation degree of each first clustering cluster at each moment, further clustering all the first clustering clusters to obtain second clustering clusters, and obtaining the tension change degree of the loom according to the position distribution of each first clustering cluster in each second clustering cluster. And performing PID control on the loom according to the deviation condition of the tension data of the loom at the current moment and the tension change degree. The control effect on the yarn tension of the high-speed loom can be improved.

Description

Technical Field

[0001] The present invention relates to the field of textile control, and in particular to a tension control method and system for a high-speed loom. Background Art

[0002] With the rapid development of the textile industry towards intelligence and efficiency, high-speed looms have put forward higher requirements for the precise control of warp yarn tension. Traditional mechanical tension devices have problems such as response lag and insufficient adjustment accuracy. They are difficult to adapt to the control requirements of instantaneous fluctuations in yarn tension in modern high-speed weaving processes, which can easily lead to quality problems such as yarn breakage and cloth defects.

[0003] In the related art, a PID controller with fixed parameters is usually used to control the tension of the loom yarn. However, due to the complex nonlinear disturbances caused by factors such as yarn material changes, feed speed fluctuations and mechanical vibrations during the weaving process of high-speed looms, and the time-varying and lagging characteristics of the tension system of high-speed looms, the existing PID controller with fixed parameters is difficult to achieve accurate and fast dynamic response, which easily causes overshoot or under-adjustment of the yarn tension, reducing the control effect of the yarn tension of the high-speed loom. Summary of the Invention

[0004] In order to solve the technical problem that the existing PID controller with fixed parameters is difficult to achieve accurate and fast dynamic response, easily causes overshoot or under-adjustment of yarn tension, and reduces the control effect of yarn tension of high-speed looms, the purpose of the present invention is to provide a tension control method and system for high-speed looms. The technical solutions adopted are as follows: The present invention provides a tension control method for a high-speed loom, the method comprising: Real-time acquisition of yarn tension data during the operation of high-speed looms; Obtaining the tension fluctuation degree of the loom at each moment based on the difference between the tension data of the loom at each moment and the overall level of the tension data at all moments, as well as the distribution of the tension data at each moment within a preset time domain at each moment; clustering all moments based on the difference in the tension fluctuation degree at each moment to obtain a plurality of first clusters; obtaining a change state value of each first cluster based on the distribution of the tension data at each moment in each first cluster and the difference in the tension fluctuation degree at each moment between the first clusters; Clustering all first clusters according to the overall level of tension data at all moments in each first cluster and the change state value to obtain a plurality of second clusters; taking any second cluster as a target second cluster, and obtaining the tension anomaly value of the target second cluster according to the position distribution of each first cluster in the target second cluster; obtaining the tension change degree of the loom according to the distribution of the tension anomaly values ​​of all second clusters; The loom is PID-controlled according to the difference between the tension data of the loom at the current moment and the overall level of the tension data at all moments, as well as the degree of tension change.

[0005] Furthermore, obtaining the tension fluctuation degree of the loom at each moment includes: The average value of the tension data of the loom at all times is taken as the overall tension value of the loom; The absolute value of the difference between the tension data of the loom at each moment and the overall tension value is used as the tension deviation value of the loom at each moment; Analyze the discrete degree of tension data at all moments within the preset time domain at each moment to obtain the local tension discrete degree at each moment; The tension deviation value and the local tension dispersion of the loom at each moment are integrated and normalized to obtain the tension fluctuation degree of the loom at each moment.

[0006] Furthermore, obtaining a plurality of first clusters includes: Taking the absolute value of the difference between the tension fluctuation levels at any two moments as a first distance measure between the two moments; All moments are clustered based on the first distance metric to obtain a plurality of first clusters.

[0007] Furthermore, obtaining the change state value of each first cluster includes: Using the PCA principal component analysis algorithm, the tension data at all moments in each first cluster are processed to obtain the principal component direction of each first cluster; Taking the average value of the tension fluctuation degree at all moments in each first cluster as the cluster center value of each first cluster; The average value of the tension fluctuation degree of the loom at all times is taken as the overall tension fluctuation value of the loom; The absolute value of the difference between the cluster center value of each first cluster and the overall tension fluctuation value is used as the tension fluctuation deviation of each first cluster; The entropy value of the tension data at all moments in each first cluster is used as the tension disorder degree of each first cluster; The principal component direction, the tension fluctuation deviation and the tension disorder of each first cluster are integrated and normalized to obtain a change state value of each first cluster.

[0008] Furthermore, obtaining a plurality of second clusters includes: The average value of the tension data at all moments in each first cluster is used as the overall tension level of each first cluster, and the two-dimensional data composed of the change state value and the overall tension level of each first cluster is used as the feature array of each first cluster; Using the Euclidean distance of the feature array between any two first clusters as a second distance metric between any two first clusters; All first clusters are clustered based on the second distance metric to obtain multiple second clusters.

[0009] Furthermore, obtaining the tension anomaly value of the target second cluster includes: The average value of the second distance measure between all any two first clusters in the target second cluster is used as the tension outlier value of the target second cluster.

[0010] Furthermore, obtaining the tension variation degree of the loom includes: The range of the tension anomaly values ​​of all the second clusters is used as the first tension variation coefficient of the loom; Analyzing the degree of dispersion of the tension abnormal values ​​of all second clusters to obtain a second tension variation coefficient of the loom; The first tension variation coefficient and the second tension variation coefficient are integrated and normalized to obtain the tension variation degree of the loom.

[0011] Furthermore, the PID control of the loom includes: Using the method for calculating the tension deviation value of the loom at each moment, obtaining the tension deviation value of the loom at the current moment; The product of the tension deviation value of the loom at the current moment and the tension variation degree of the loom is used as the adjustment coefficient of the loom at the current moment; Based on the adjustment coefficient of the loom at the current moment, real-time PID control is performed on the line feed speed of the loom.

[0012] Furthermore, the real-time PID control of the line feed speed of the loom based on the adjustment coefficient of the loom at the current moment includes: Based on the calculation formula of the control input signal, the control input signal of the PID controller at the current moment is obtained. The calculation formula of the control input signal is: in, Represents the control input signal of the PID controller at the current moment; represents the adjustment coefficient of the loom at the current moment; Indicates the thread feeding speed of the loom at the current moment; Indicates the integral gain parameter of the PID controller; Indicates systematic error; Represents the differential gain parameter of the PID controller; The control input signal is input into the PID controller, and the PID controller controls the line feed speed of the loom at the current moment.

[0013] The present invention also proposes a tension control system for a high-speed loom, the system comprising a memory, a processor, and a computer program stored in the memory and runnable on the processor, wherein the processor implements any one of the steps of a tension control method for a high-speed loom when executing the computer program.

[0014] The present invention has the following beneficial effects: The present invention takes into account that the existing PID controller with fixed parameters is difficult to achieve accurate and fast dynamic response, which easily causes overshoot or insufficient adjustment of yarn tension, and reduces the control effect of yarn tension of high-speed looms. Therefore, the yarn tension data of the high-speed loom during operation is first collected in real time. Taking into account that insufficient tension control on the yarn of the high-speed loom will cause fluctuations in the tension of the yarn, the degree of tension fluctuation on the yarn of the loom can be reflected in real time by the obtained tension fluctuation degree; and then the moments with similar tension fluctuations are divided into the same first cluster, and the obtained change state value is used to reflect the change status of the tension data at each moment in each first cluster. At the same time, in order to be able to Accurately describe the changes in yarn tension of the loom during the entire operation process. The present invention further clusters all first clusters, and divides situations with similar tension changes during the operation of the loom into the same second cluster. The obtained tension anomaly value reflects the degree of inconsistency between the tension characteristics and tension change characteristics between each first cluster in the target second cluster, and then reflects the degree of change in yarn tension of the high-speed loom during the entire operation process through the tension change degree. Based on the difference between the tension data of the loom at the current moment and the overall level of the tension data at all moments, and combined with the tension change degree, PID control is performed on the loom, thereby improving the control effect of the yarn tension of the high-speed loom. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] In order to more clearly illustrate the technical solutions and advantages of the embodiments of the present invention or the prior art, the following briefly introduces the drawings required for use in the embodiments or the prior art descriptions. Obviously, the drawings described below are only some embodiments of the present invention. For ordinary technicians in this field, other drawings can be obtained based on these drawings without paying any creative work.

[0016] Figure 1 A flow chart of a tension control method for a high-speed loom provided by one embodiment of the present invention. DETAILED DESCRIPTION

[0017] To further illustrate the technical means and effectiveness of the present invention in achieving its intended objectives, the following, in conjunction with the accompanying drawings and preferred embodiments, describes in detail a tension control method and system for a high-speed loom according to the present invention, including its specific implementation, structure, features, and effectiveness. In the following description, references to "one embodiment" or "another embodiment" do not necessarily refer to the same embodiment. Furthermore, specific features, structures, or characteristics of one or more embodiments may be combined in any suitable manner.

[0018] Unless defined otherwise, 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 invention belongs.

[0019] The specific scheme of the tension control method and system for a high-speed loom provided by the present invention is described in detail below with reference to the accompanying drawings.

[0020] See also Figure 1 , which shows a flow chart of a tension control method for a high-speed loom provided by one embodiment of the present invention, the method comprising: Step S1: obtaining yarn tension data of a high-speed loom in real time during operation.

[0021] During the operation of high-speed looms, stable tension can prevent yarn breakage due to excessive tension or accumulation due to excessive looseness, thereby reducing downtime and improving production efficiency. At the same time, reasonable tension control can ensure the uniformity of the fabric structure, avoid defects such as wrinkles, weft fringes or deformation, and ensure that the appearance and feel of the finished fabric meet the standards. In addition, good tension management can also reduce wear on mechanical parts, extend the service life of the equipment, and adapt to the needs of different materials and complex processes, such as the weaving of stretch fabrics or high-density fabrics.

[0022] In an embodiment of the present invention, a high-precision strain gauge tension sensor or a piezoelectric sensor is first installed on a yarn guide roller or a tension roller of a high-speed loom yarn path, and the sensor is used to collect yarn tension data of the high-speed loom in real time during operation.

[0023] Step S2: According to the difference between the tension data of the loom at each moment and the overall level of the tension data of all moments, as well as the distribution of the tension data at each moment within the preset time domain of each moment, the tension fluctuation degree of the loom is obtained; according to the difference in the tension fluctuation degree at each moment, all moments are clustered to obtain multiple first clusters; according to the distribution of the tension data at each moment in each first cluster, as well as the difference in the tension fluctuation degree at each moment between the first clusters, the change state value of each first cluster is obtained.

[0024] In the tension control of high-speed looms, PID control systems are widely used because of their simple structure and good adjustment performance. The system can quickly respond to tension fluctuations, quickly reduce deviations through the proportional link, eliminate steady-state errors through the integral link, and suppress overshoot through the differential link, thereby ensuring yarn tension stability and improving fabric quality. However, PID control also has some shortcomings: on the one hand, traditional PID parameters usually rely on experience to set, and it may be difficult to achieve optimal control effects under complex working conditions or high dynamic change conditions (such as frequent start and stop, yarn mutation, etc.). On the other hand, PID control has limited adaptability to nonlinear and time-varying systems. When the loom speed is extremely high or the tension disturbance is large, adjustment lag or oscillation may occur. When high-speed weaving is in operation, the control may fail. When the tension on the machine yarn is not adequately controlled, the tension on the yarn will fluctuate. Therefore, the embodiment of the present invention first analyzes the difference between the tension data of the loom at each moment and the overall level of the tension data at all moments, as well as the distribution of the tension data at each moment within the preset time domain of each moment. The obtained tension fluctuation degree reflects the degree of tension fluctuation on the yarn of the loom in real time. The length range of the preset time domain is usually 10~30. In the embodiment of the present invention, the length of the preset time domain is set to 15, that is, the preset time domain at a certain moment includes 14 other moments closest to the moment and the moment itself. The specific length of the preset time domain can also be set by the implementer according to the specific implementation scenario, which is not limited here.

[0025] Preferably, in one embodiment of the present invention, the method for obtaining the tension fluctuation degree of the loom at each moment specifically includes: The average value of the tension data of the loom at all times is taken as the overall tension value of the loom, and the overall tension value reflects the overall level of yarn tension of the high-speed loom during the entire operation process.

[0026] The absolute value of the difference between the tension data of the loom at each moment and the overall tension value is taken as the tension deviation value of the loom at each moment. The larger the tension deviation value at a certain moment, the more the tension of the yarn at that moment deviates from the overall tension level, and further the greater the instantaneous fluctuation of the yarn tension at that moment.

[0027] The discrete degree of the tension data at all moments in the preset time domain at each moment is analyzed to obtain the local tension discreteness at each moment. The larger the local tension discreteness, the more obvious the tension fluctuation of the loom yarn in the local short time at that moment.

[0028] In an embodiment of the present invention, the standard deviation or variance of the tension data at all moments in the preset time domain of each moment can be used as the local tension discreteness at each moment, thereby realizing the analysis of the discreteness of the tension data at all moments in the preset time domain of each moment. This is not limited here, and the same method can also be used in subsequent steps to realize the analysis of the discreteness of the data.

[0029] Then, the tension deviation value and local tension dispersion of the loom at each moment are integrated and normalized, and the calculation results are limited to range, thereby obtaining the tension fluctuation degree of the loom at each moment.

[0030] In an embodiment of the present invention, the combination of the tension deviation value and the local tension dispersion of the loom at each moment can be achieved by calculating the sum or product of the two, which is not limited here, and the subsequent steps of the comprehensive processing of two or more data can also be achieved using the same method.

[0031] In one embodiment of the present invention, the normalization processing can be specifically, for example, maximum and minimum value normalization processing, and the normalization in subsequent steps can all adopt maximum and minimum value normalization processing. In other embodiments of the present invention, other normalization methods can be selected according to the specific range of values, or activation functions and hyperbolic tangent functions can be used to implement normalization processing, which will not be repeated or limited.

[0032] As an example, in one embodiment of the present invention, the expression for the tension fluctuation degree of the loom at each moment can be specifically, for example, as follows: in, Indicates that the loom is The degree of tension fluctuation at each moment; Indicates that the loom is Tension data at each moment; Indicates the overall tension value of the loom; Indicates that the loom is Tension deviation value at each moment; Indicates the The standard deviation of the tension data at all moments in the preset time domain, that is, The local tension dispersion at each moment; Represents the normalization function, used for normalization processing.

[0033] During the entire operation process of the high-speed loom, there are moments with similar tension fluctuations. Therefore, the embodiment of the present invention clusters all moments according to the differences in the degree of tension fluctuation at each moment to obtain multiple first clusters, thereby realizing the division of moments with similar tension fluctuations into the same cluster. Subsequently, the tension data and tension fluctuation degree at each moment in each first cluster can be analyzed, thereby accurately analyzing the changes in the tension data at each moment in each first cluster.

[0034] Preferably, in one embodiment of the present invention, the method for obtaining the plurality of first clusters specifically includes: The absolute value of the difference in tension fluctuation between any two moments is used as the first distance metric between any two moments. Based on the first distance metric, all moments are clustered to obtain multiple first clusters. In one embodiment of the present invention, the existing K-meams clustering algorithm can be used to implement the clustering operation, and the number of clusters can be determined using the existing elbow method. In other embodiments of the present invention, hierarchical clustering algorithms or other clustering algorithms can also be used to implement clustering operations, which are not limited here.

[0035] During the spinning process of a high-speed loom, there are many factors that cause changes in the yarn tension. For example, uneven braking of the warp beam, delayed response of the servo motor, or mechanical wear will cause fluctuations in the feed speed, which will directly lead to changes in the yarn tension. The changes in the tension data at each moment in different first clusters are different, so it is necessary to analyze each first cluster. The embodiment of the present invention obtains the change state value of each first cluster by analyzing the distribution of tension data at each moment in each first cluster and the difference in the degree of tension fluctuation at each moment between the first clusters. The change state value reflects the tension change at each moment in each first cluster.

[0036] Preferably, in one embodiment of the present invention, the method for obtaining the change state value of each first cluster specifically includes: First, the PCA principal component analysis algorithm is used to process the tension data at all moments in each first cluster to obtain the principal component direction of each first cluster. The larger the principal component direction of a first cluster, the greater the change trend of the tension data at each moment in the first cluster. The PCA principal component analysis algorithm is a technical means well known to those skilled in the art and will not be elaborated here.

[0037] The average value of the tension fluctuation degree at all moments in each first cluster is taken as the cluster center value of each first cluster, the average value of the tension fluctuation degree of the loom at all moments is taken as the overall tension fluctuation value of the loom, and the absolute value of the difference between the cluster center value of each first cluster and the overall tension fluctuation value is taken as the tension fluctuation deviation of each first cluster. The greater the tension fluctuation deviation of a first cluster, the more the tension fluctuation characteristics at each moment in the first cluster deviate from the overall level of tension fluctuation of the loom during the entire operation process.

[0038] The entropy value of the tension data at all moments in each first cluster is used as the tension disorder of each first cluster. The greater the tension disorder, the more inconsistent the tension data at each moment in the first cluster. The calculation of the entropy value of the data is a technical means well known to those skilled in the art and will not be elaborated here.

[0039] Then, the principal component direction, tension fluctuation deviation and tension disorder of each first cluster are integrated and normalized, and the calculation results are limited to range, thereby obtaining the change state value of each first cluster.

[0040] As an example, in one embodiment of the present invention, the expression of the change state value of each first cluster may be specifically, for example, as follows: in, Indicates the The change state value of the first cluster; Indicates the The principal component direction of the first cluster; Indicates the The cluster center value of the first cluster; Indicates the overall tension fluctuation value of the loom; Indicates the The tension fluctuation deviation of the first cluster; Indicates the The tension disorder of the first cluster; Represents the normalization function, used for normalization processing.

[0041] At this point, the analysis of the change state of the tension data at each moment in each first cluster is completed.

[0042] Step S3: Cluster all first clusters according to the overall level and change state value of the tension data at all moments in each first cluster to obtain multiple second clusters; take any second cluster as the target second cluster, and obtain the tension anomaly value of the target second cluster according to the position distribution of each first cluster in the target second cluster; obtain the tension change degree of the loom according to the distribution of the tension anomaly values ​​of all second clusters.

[0043] Since different first clusters represent the changing patterns of yarn tension during the operation of the high-speed loom, in order to describe the changes in yarn tension during the operation of the high-speed loom, it is necessary to further classify the first clusters, so as to determine the degree of adjustment of the feed speed of the high-speed loom and improve the control effect of the yarn tension of the high-speed loom. The embodiment of the present invention first clusters all first clusters according to the overall level and change state value of the tension data at all moments in each first cluster to obtain multiple second clusters. The first clusters in the same second cluster have similar tension change characteristics and similar tension characteristics.

[0044] Preferably, in one embodiment of the present invention, the method for obtaining the plurality of second clusters specifically includes: The average value of the tension data at all moments in each first cluster is used as the overall tension level of each first cluster, and the two-dimensional data composed of the change state value and the overall tension level of each first cluster is used as the feature array of each first cluster. At this time, each first cluster can be regarded as a data point, and each data point is distributed in the two-dimensional plane composed of the change state value and the overall tension level.

[0045] The Euclidean distance of the feature array between any two first clusters is used as the second distance metric between any two first clusters. Based on the second distance metric, all first clusters are clustered to obtain multiple second clusters. In one embodiment of the present invention, the existing DBSCAN clustering algorithm or other clustering algorithms can be selected to implement clustering processing, which is not limited or elaborated here.

[0046] Then, any second cluster is analyzed and any second cluster is used as the target second cluster. For the target second cluster, if the first clusters in the target second cluster are regarded as data points, the more concentrated the position distribution of the first clusters in the target second cluster is, the more similar the tension change characteristics and tension characteristics between the first clusters in the target second cluster are. On the contrary, the more discrete the position distribution of the first cluster in the target second cluster is, the greater the difference in the tension change characteristics and tension characteristics between the first clusters in the target second cluster, which further indicates that the abnormality of the yarn tension shown by the target second cluster is more obvious. Therefore, the tension abnormality value of the target second cluster can be obtained according to the position distribution of the first clusters in the target second cluster. Subsequently, the tension abnormality value of each second cluster can be combined to analyze the degree of tension change of the high-speed loom during the entire operation process, thereby improving the effect of yarn tension control.

[0047] Preferably, in one embodiment of the present invention, the method for obtaining the tension anomaly value of the target second cluster specifically includes: The average value of the second distance metric between all any two first clusters in the target second cluster is taken as the tension outlier value of the target second cluster.

[0048] The tension anomaly value of each second cluster can be obtained by the same method as above. The greater the degree of dispersion of the tension anomaly value distribution of each second cluster, the greater the degree of change in the yarn tension of the high-speed loom during operation. In the subsequent PID control process, the yarn feed speed needs to be adjusted to a greater extent to ensure the stability of the yarn tension. Therefore, the distribution characteristics of the tension anomaly values ​​of all second clusters can be analyzed. The degree of tension change obtained can reflect the degree of change in the yarn tension of the high-speed loom during the entire operation. Subsequently, the feed speed of the loom can be controlled based on the tension change degree, thereby ensuring the stability of the yarn tension and improving the control effect of the yarn tension.

[0049] Preferably, in one embodiment of the present invention, the method for obtaining the tension variation degree of a loom specifically includes: The range of the tension anomaly values ​​of all the second clusters is used as the first tension variation coefficient of the loom. The discrete degree of the tension anomaly values ​​of all the second clusters is analyzed to obtain the second tension variation coefficient of the loom. The larger the first tension variation coefficient and the second tension coefficient, the greater the degree of tension variation during the entire operation of the loom.

[0050] Then, the first tension variation coefficient and the second tension variation coefficient are combined and normalized, and the calculation result is limited to range, thereby obtaining the degree of tension change of the loom.

[0051] As an example, in one embodiment of the present invention, the expression for the degree of tension change of the loom may be specifically, for example, as follows: in, Indicates the degree of tension change in the loom; Indicates the maximum value of the tension anomaly values ​​of all the second clusters; Represents the minimum value of the tension anomaly of all second clusters; Represents the range of tension anomalies of all second clusters, that is, the first tension variation coefficient of the loom; represents the standard deviation of the tension anomalies of all second clusters, that is, the second tension variation coefficient of the loom; Represents the hyperbolic tangent function, which is used for normalization.

[0052] Step S4: PID control is performed on the loom according to the difference between the tension data of the loom at the current moment and the overall level of the tension data at all moments, as well as the degree of tension change.

[0053] The feed speed of a high-speed loom is a key factor affecting the yarn tension. When the feed speed of a high-speed loom is unstable during operation, the yarn tension will fluctuate instantaneously. Therefore, in order to achieve more accurate real-time adjustment of the yarn tension of the high-speed loom, an embodiment of the present invention performs more accurate PID control of the feed speed of the loom based on the difference between the tension data of the loom at the current moment and the overall level of the tension data at all moments, and in combination with the degree of tension change of the loom, thereby improving the control effect of the yarn tension.

[0054] Preferably, in one embodiment of the present invention, the method for performing PID control on the feed speed of a loom specifically includes: The tension deviation value of the loom at the current moment is obtained by using the calculation method of the tension deviation value of the loom at each moment. The specific calculation method is: the absolute value of the difference between the tension data of the loom at the current moment and the overall tension value is used as the tension deviation value of the loom at the current moment.

[0055] The product of the tension deviation value of the loom at the current moment and the tension change degree of the loom is used as the adjustment coefficient of the loom at the current moment. Based on the adjustment coefficient of the loom at the current moment, real-time PID control is performed on the feed speed of the loom.

[0056] Preferably, in one embodiment of the present invention, the method for performing PID control on the feed speed of a loom further comprises: Based on the calculation formula of the control input signal, the control input signal of the PID controller at the current moment is obtained. The calculation formula of the control input signal is: in, Represents the control input signal of the PID controller at the current moment; Indicates the adjustment coefficient of the loom at the current moment; Indicates the thread feeding speed of the loom at the current moment; Indicates the integral gain parameter of the PID controller; Represents the system error, which can be obtained by the PID controller itself; Represents the differential gain parameter of the PID controller; The control input signal is input into the PID controller, and the PID controller controls the feeding speed of the loom at the current moment. After the feeding speed of the high-speed loom is effectively controlled, the tension of the yarn will also remain stable.

[0057] One embodiment of the present invention provides a tension control system for a high-speed loom, the system comprising a memory, a processor, and a computer program, wherein the memory is used to store the corresponding computer program, the processor is used to run the corresponding computer program, and when the computer program runs in the processor, it can implement the method described in steps S1 to S4.

[0058] It should be noted that the order in which the embodiments of the present invention are described above is for illustrative purposes only and does not necessarily represent the superiority or inferiority of the embodiments. The processes depicted in the accompanying drawings do not necessarily require the specific order or sequential order shown to achieve the desired results. In certain embodiments, multitasking and parallel processing are also possible or may be advantageous.

[0059] The various embodiments in this specification are described in a progressive manner, and the same or similar parts between the various embodiments can be referred to each other. Each embodiment focuses on the differences from other embodiments.

Claims

1. A tension control method for a high-speed loom, characterized in that: The method comprises: Real-time acquisition of yarn tension data during the operation of high-speed looms; Obtaining the tension fluctuation degree of the loom at each moment based on the difference between the tension data of the loom at each moment and the overall level of the tension data at all moments, as well as the distribution of the tension data at each moment within a preset time domain at each moment; clustering all moments based on the difference in the tension fluctuation degree at each moment to obtain a plurality of first clusters; obtaining a change state value of each first cluster based on the distribution of the tension data at each moment in each first cluster and the difference in the tension fluctuation degree at each moment between the first clusters; Clustering all first clusters according to the overall level of tension data at all moments in each first cluster and the change state value to obtain a plurality of second clusters; taking any second cluster as a target second cluster, and obtaining the tension anomaly value of the target second cluster according to the position distribution of each first cluster in the target second cluster; obtaining the tension change degree of the loom according to the distribution of the tension anomaly values ​​of all second clusters; The loom is PID-controlled according to the difference between the tension data of the loom at the current moment and the overall level of the tension data at all moments, as well as the degree of tension change.

2. The tension control method for a high-speed loom according to claim 1, characterized in that: The obtaining of the tension fluctuation degree of the loom at each moment comprises: The average value of the tension data of the loom at all times is taken as the overall tension value of the loom; The absolute value of the difference between the tension data of the loom at each moment and the overall tension value is used as the tension deviation value of the loom at each moment; Analyze the discrete degree of tension data at all moments within the preset time domain at each moment to obtain the local tension discrete degree at each moment; The tension deviation value and the local tension dispersion of the loom at each moment are integrated and normalized to obtain the tension fluctuation degree of the loom at each moment.

3. The tension control method for a high-speed loom according to claim 1, characterized in that: The obtaining of a plurality of first clusters comprises: Taking the absolute value of the difference between the tension fluctuation levels at any two moments as a first distance measure between the two moments; All moments are clustered based on the first distance metric to obtain a plurality of first clusters.

4. The tension control method for a high-speed loom according to claim 1, characterized in that: The obtaining of the change state value of each first cluster includes: Using the PCA principal component analysis algorithm, the tension data at all moments in each first cluster are processed to obtain the principal component direction of each first cluster; Taking the average value of the tension fluctuation degree at all moments in each first cluster as the cluster center value of each first cluster; The average value of the tension fluctuation degree of the loom at all times is taken as the overall tension fluctuation value of the loom; The absolute value of the difference between the cluster center value of each first cluster and the overall tension fluctuation value is used as the tension fluctuation deviation of each first cluster; The entropy value of the tension data at all moments in each first cluster is used as the tension disorder degree of each first cluster; The principal component direction, the tension fluctuation deviation and the tension disorder of each first cluster are integrated and normalized to obtain a change state value of each first cluster.

5. The tension control method for a high-speed loom according to claim 1, characterized in that: The obtaining of a plurality of second clusters comprises: The average value of the tension data at all moments in each first cluster is used as the overall tension level of each first cluster, and the two-dimensional data composed of the change state value and the overall tension level of each first cluster is used as the feature array of each first cluster; Using the Euclidean distance of the feature array between any two first clusters as a second distance metric between any two first clusters; All first clusters are clustered based on the second distance metric to obtain multiple second clusters.

6. The tension control method for a high-speed loom according to claim 5, characterized in that: The obtaining of the tension anomaly value of the target second cluster comprises: The average value of the second distance measure between all any two first clusters in the target second cluster is used as the tension outlier value of the target second cluster.

7. The tension control method for a high-speed loom according to claim 1, characterized in that: The obtaining of the tension variation degree of the loom comprises: The range of the tension anomaly values ​​of all the second clusters is used as the first tension variation coefficient of the loom; Analyzing the degree of dispersion of the tension abnormal values ​​of all second clusters to obtain a second tension variation coefficient of the loom; The first tension variation coefficient and the second tension variation coefficient are integrated and normalized to obtain the tension variation degree of the loom.

8. The tension control method for a high-speed loom according to claim 2, characterized in that: The PID control of the loom comprises: Using the method for calculating the tension deviation value of the loom at each moment, obtaining the tension deviation value of the loom at the current moment; The product of the tension deviation value of the loom at the current moment and the tension variation degree of the loom is used as the adjustment coefficient of the loom at the current moment; Based on the adjustment coefficient of the loom at the current moment, real-time PID control is performed on the line feed speed of the loom.

9. The tension control method for a high-speed loom according to claim 8, characterized in that: The real-time PID control of the line feed speed of the loom based on the adjustment coefficient of the loom at the current moment includes: Based on the calculation formula of the control input signal, the control input signal of the PID controller at the current moment is obtained. The calculation formula of the control input signal is: in, Represents the control input signal of the PID controller at the current moment; represents the adjustment coefficient of the loom at the current moment; Indicates the thread feeding speed of the loom at the current moment; Indicates the integral gain parameter of the PID controller; Indicates systematic error; Represents the differential gain parameter of the PID controller; The control input signal is input into the PID controller, and the PID controller controls the line feed speed of the loom at the current moment.

10. A tension control system for a high-speed loom, the system comprising a memory, a processor, and a computer program stored in the memory and executable on the processor, wherein: When the processor executes the computer program, the steps of the method according to any one of claims 1 to 9 are implemented.

Images